Superposition-Based Analysis of First-Order Probabilistic Timed Automata

NASA Astrophysics Data System (ADS)

Fietzke, Arnaud; Hermanns, Holger; Weidenbach, Christoph

This paper discusses the analysis of first-order probabilistic timed automata (FPTA) by a combination of hierarchic first-order superposition-based theorem proving and probabilistic model checking. We develop the overall semantics of FPTAs and prove soundness and completeness of our method for reachability properties. Basically, we decompose FPTAs into their time plus first-order logic aspects on the one hand, and their probabilistic aspects on the other hand. Then we exploit the time plus first-order behavior by hierarchic superposition over linear arithmetic. The result of this analysis is the basis for the construction of a reachability equivalent (to the original FPTA) probabilistic timed automaton to which probabilistic model checking is finally applied. The hierarchic superposition calculus required for the analysis is sound and complete on the first-order formulas generated from FPTAs. It even works well in practice. We illustrate the potential behind it with a real-life DHCP protocol example, which we analyze by means of tool chain support.

Probabilistic modeling of bifurcations in single-cell gene expression data using a Bayesian mixture of factor analyzers.

PubMed

Campbell, Kieran R; Yau, Christopher

2017-03-15

Modeling bifurcations in single-cell transcriptomics data has become an increasingly popular field of research. Several methods have been proposed to infer bifurcation structure from such data, but all rely on heuristic non-probabilistic inference. Here we propose the first generative, fully probabilistic model for such inference based on a Bayesian hierarchical mixture of factor analyzers. Our model exhibits competitive performance on large datasets despite implementing full Markov-Chain Monte Carlo sampling, and its unique hierarchical prior structure enables automatic determination of genes driving the bifurcation process. We additionally propose an Empirical-Bayes like extension that deals with the high levels of zero-inflation in single-cell RNA-seq data and quantify when such models are useful. We apply or model to both real and simulated single-cell gene expression data and compare the results to existing pseudotime methods. Finally, we discuss both the merits and weaknesses of such a unified, probabilistic approach in the context practical bioinformatics analyses.

A range of complex probabilistic models for RNA secondary structure prediction that includes the nearest-neighbor model and more.

PubMed

Rivas, Elena; Lang, Raymond; Eddy, Sean R

2012-02-01

The standard approach for single-sequence RNA secondary structure prediction uses a nearest-neighbor thermodynamic model with several thousand experimentally determined energy parameters. An attractive alternative is to use statistical approaches with parameters estimated from growing databases of structural RNAs. Good results have been reported for discriminative statistical methods using complex nearest-neighbor models, including CONTRAfold, Simfold, and ContextFold. Little work has been reported on generative probabilistic models (stochastic context-free grammars [SCFGs]) of comparable complexity, although probabilistic models are generally easier to train and to use. To explore a range of probabilistic models of increasing complexity, and to directly compare probabilistic, thermodynamic, and discriminative approaches, we created TORNADO, a computational tool that can parse a wide spectrum of RNA grammar architectures (including the standard nearest-neighbor model and more) using a generalized super-grammar that can be parameterized with probabilities, energies, or arbitrary scores. By using TORNADO, we find that probabilistic nearest-neighbor models perform comparably to (but not significantly better than) discriminative methods. We find that complex statistical models are prone to overfitting RNA structure and that evaluations should use structurally nonhomologous training and test data sets. Overfitting has affected at least one published method (ContextFold). The most important barrier to improving statistical approaches for RNA secondary structure prediction is the lack of diversity of well-curated single-sequence RNA secondary structures in current RNA databases.

A range of complex probabilistic models for RNA secondary structure prediction that includes the nearest-neighbor model and more

PubMed Central

Rivas, Elena; Lang, Raymond; Eddy, Sean R.

2012-01-01

The standard approach for single-sequence RNA secondary structure prediction uses a nearest-neighbor thermodynamic model with several thousand experimentally determined energy parameters. An attractive alternative is to use statistical approaches with parameters estimated from growing databases of structural RNAs. Good results have been reported for discriminative statistical methods using complex nearest-neighbor models, including CONTRAfold, Simfold, and ContextFold. Little work has been reported on generative probabilistic models (stochastic context-free grammars [SCFGs]) of comparable complexity, although probabilistic models are generally easier to train and to use. To explore a range of probabilistic models of increasing complexity, and to directly compare probabilistic, thermodynamic, and discriminative approaches, we created TORNADO, a computational tool that can parse a wide spectrum of RNA grammar architectures (including the standard nearest-neighbor model and more) using a generalized super-grammar that can be parameterized with probabilities, energies, or arbitrary scores. By using TORNADO, we find that probabilistic nearest-neighbor models perform comparably to (but not significantly better than) discriminative methods. We find that complex statistical models are prone to overfitting RNA structure and that evaluations should use structurally nonhomologous training and test data sets. Overfitting has affected at least one published method (ContextFold). The most important barrier to improving statistical approaches for RNA secondary structure prediction is the lack of diversity of well-curated single-sequence RNA secondary structures in current RNA databases. PMID:22194308

Probabilistic Seismic Risk Model for Western Balkans

NASA Astrophysics Data System (ADS)

Stejskal, Vladimir; Lorenzo, Francisco; Pousse, Guillaume; Radovanovic, Slavica; Pekevski, Lazo; Dojcinovski, Dragi; Lokin, Petar; Petronijevic, Mira; Sipka, Vesna

2010-05-01

A probabilistic seismic risk model for insurance and reinsurance purposes is presented for an area of Western Balkans, covering former Yugoslavia and Albania. This territory experienced many severe earthquakes during past centuries producing significant damage to many population centres in the region. The highest hazard is related to external Dinarides, namely to the collision zone of the Adriatic plate. The model is based on a unified catalogue for the region and a seismic source model consisting of more than 30 zones covering all the three main structural units - Southern Alps, Dinarides and the south-western margin of the Pannonian Basin. A probabilistic methodology using Monte Carlo simulation was applied to generate the hazard component of the model. Unique set of damage functions based on both loss experience and engineering assessments is used to convert the modelled ground motion severity into the monetary loss.

Climatological attribution of wind power ramp events in East Japan and their probabilistic forecast based on multi-model ensembles downscaled by analog ensemble using self-organizing maps

NASA Astrophysics Data System (ADS)

Ohba, Masamichi; Nohara, Daisuke; Kadokura, Shinji

2016-04-01

Severe storms or other extreme weather events can interrupt the spin of wind turbines in large scale that cause unexpected "wind ramp events". In this study, we present an application of self-organizing maps (SOMs) for climatological attribution of the wind ramp events and their probabilistic prediction. The SOM is an automatic data-mining clustering technique, which allows us to summarize a high-dimensional data space in terms of a set of reference vectors. The SOM is applied to analyze and connect the relationship between atmospheric patterns over Japan and wind power generation. SOM is employed on sea level pressure derived from the JRA55 reanalysis over the target area (Tohoku region in Japan), whereby a two-dimensional lattice of weather patterns (WPs) classified during the 1977-2013 period is obtained. To compare with the atmospheric data, the long-term wind power generation is reconstructed by using a high-resolution surface observation network AMeDAS (Automated Meteorological Data Acquisition System) in Japan. Our analysis extracts seven typical WPs, which are linked to frequent occurrences of wind ramp events. Probabilistic forecasts to wind power generation and ramps are conducted by using the obtained SOM. The probability are derived from the multiple SOM lattices based on the matching of output from TIGGE multi-model global forecast to the WPs on the lattices. Since this method effectively takes care of the empirical uncertainties from the historical data, wind power generation and ramp is probabilistically forecasted from the forecasts of global models. The predictability skill of the forecasts for the wind power generation and ramp events show the relatively good skill score under the downscaling technique. It is expected that the results of this study provides better guidance to the user community and contribute to future development of system operation model for the transmission grid operator.

Probabilistic Models and Generative Neural Networks: Towards an Unified Framework for Modeling Normal and Impaired Neurocognitive Functions

PubMed Central

Testolin, Alberto; Zorzi, Marco

2016-01-01

Connectionist models can be characterized within the more general framework of probabilistic graphical models, which allow to efficiently describe complex statistical distributions involving a large number of interacting variables. This integration allows building more realistic computational models of cognitive functions, which more faithfully reflect the underlying neural mechanisms at the same time providing a useful bridge to higher-level descriptions in terms of Bayesian computations. Here we discuss a powerful class of graphical models that can be implemented as stochastic, generative neural networks. These models overcome many limitations associated with classic connectionist models, for example by exploiting unsupervised learning in hierarchical architectures (deep networks) and by taking into account top-down, predictive processing supported by feedback loops. We review some recent cognitive models based on generative networks, and we point out promising research directions to investigate neuropsychological disorders within this approach. Though further efforts are required in order to fill the gap between structured Bayesian models and more realistic, biophysical models of neuronal dynamics, we argue that generative neural networks have the potential to bridge these levels of analysis, thereby improving our understanding of the neural bases of cognition and of pathologies caused by brain damage. PMID:27468262

Probabilistic Models and Generative Neural Networks: Towards an Unified Framework for Modeling Normal and Impaired Neurocognitive Functions.

PubMed

Testolin, Alberto; Zorzi, Marco

2016-01-01

Connectionist models can be characterized within the more general framework of probabilistic graphical models, which allow to efficiently describe complex statistical distributions involving a large number of interacting variables. This integration allows building more realistic computational models of cognitive functions, which more faithfully reflect the underlying neural mechanisms at the same time providing a useful bridge to higher-level descriptions in terms of Bayesian computations. Here we discuss a powerful class of graphical models that can be implemented as stochastic, generative neural networks. These models overcome many limitations associated with classic connectionist models, for example by exploiting unsupervised learning in hierarchical architectures (deep networks) and by taking into account top-down, predictive processing supported by feedback loops. We review some recent cognitive models based on generative networks, and we point out promising research directions to investigate neuropsychological disorders within this approach. Though further efforts are required in order to fill the gap between structured Bayesian models and more realistic, biophysical models of neuronal dynamics, we argue that generative neural networks have the potential to bridge these levels of analysis, thereby improving our understanding of the neural bases of cognition and of pathologies caused by brain damage.

Approximate probabilistic cellular automata for the dynamics of single-species populations under discrete logisticlike growth with and without weak Allee effects.

PubMed

Mendonça, J Ricardo G; Gevorgyan, Yeva

2017-05-01

We investigate one-dimensional elementary probabilistic cellular automata (PCA) whose dynamics in first-order mean-field approximation yields discrete logisticlike growth models for a single-species unstructured population with nonoverlapping generations. Beginning with a general six-parameter model, we find constraints on the transition probabilities of the PCA that guarantee that the ensuing approximations make sense in terms of population dynamics and classify the valid combinations thereof. Several possible models display a negative cubic term that can be interpreted as a weak Allee factor. We also investigate the conditions under which a one-parameter PCA derived from the more general six-parameter model can generate valid population growth dynamics. Numerical simulations illustrate the behavior of some of the PCA found.

Control of Stochastic Master Equation Models of Genetic Regulatory Networks by Approximating Their Average Behavior

NASA Astrophysics Data System (ADS)

Umut Caglar, Mehmet; Pal, Ranadip

2010-10-01

The central dogma of molecular biology states that ``information cannot be transferred back from protein to either protein or nucleic acid.'' However, this assumption is not exactly correct in most of the cases. There are a lot of feedback loops and interactions between different levels of systems. These types of interactions are hard to analyze due to the lack of data in the cellular level and probabilistic nature of interactions. Probabilistic models like Stochastic Master Equation (SME) or deterministic models like differential equations (DE) can be used to analyze these types of interactions. SME models based on chemical master equation (CME) can provide detailed representation of genetic regulatory system, but their use is restricted by the large data requirements and computational costs of calculations. The differential equations models on the other hand, have low calculation costs and much more adequate to generate control procedures on the system; but they are not adequate to investigate the probabilistic nature of interactions. In this work the success of the mapping between SME and DE is analyzed, and the success of a control policy generated by DE model with respect to SME model is examined. Index Terms--- Stochastic Master Equation models, Differential Equation Models, Control Policy Design, Systems biology

A Probabilistic Account of Exemplar and Category Generation

ERIC Educational Resources Information Center

Jern, Alan; Kemp, Charles

2013-01-01

People are capable of imagining and generating new category exemplars and categories. This ability has not been addressed by previous models of categorization, most of which focus on classifying category exemplars rather than generating them. We develop a formal account of exemplar and category generation which proposes that category knowledge is…

The meta-Gaussian Bayesian Processor of forecasts and associated preliminary experiments

NASA Astrophysics Data System (ADS)

Chen, Fajing; Jiao, Meiyan; Chen, Jing

2013-04-01

Public weather services are trending toward providing users with probabilistic weather forecasts, in place of traditional deterministic forecasts. Probabilistic forecasting techniques are continually being improved to optimize available forecasting information. The Bayesian Processor of Forecast (BPF), a new statistical method for probabilistic forecast, can transform a deterministic forecast into a probabilistic forecast according to the historical statistical relationship between observations and forecasts generated by that forecasting system. This technique accounts for the typical forecasting performance of a deterministic forecasting system in quantifying the forecast uncertainty. The meta-Gaussian likelihood model is suitable for a variety of stochastic dependence structures with monotone likelihood ratios. The meta-Gaussian BPF adopting this kind of likelihood model can therefore be applied across many fields, including meteorology and hydrology. The Bayes theorem with two continuous random variables and the normal-linear BPF are briefly introduced. The meta-Gaussian BPF for a continuous predictand using a single predictor is then presented and discussed. The performance of the meta-Gaussian BPF is tested in a preliminary experiment. Control forecasts of daily surface temperature at 0000 UTC at Changsha and Wuhan stations are used as the deterministic forecast data. These control forecasts are taken from ensemble predictions with a 96-h lead time generated by the National Meteorological Center of the China Meteorological Administration, the European Centre for Medium-Range Weather Forecasts, and the US National Centers for Environmental Prediction during January 2008. The results of the experiment show that the meta-Gaussian BPF can transform a deterministic control forecast of surface temperature from any one of the three ensemble predictions into a useful probabilistic forecast of surface temperature. These probabilistic forecasts quantify the uncertainty of the control forecast; accordingly, the performance of the probabilistic forecasts differs based on the source of the underlying deterministic control forecasts.

Effects of varying the step particle distribution on a probabilistic transport model

NASA Astrophysics Data System (ADS)

Bouzat, S.; Farengo, R.

2005-12-01

The consequences of varying the step particle distribution on a probabilistic transport model, which captures the basic features of transport in plasmas and was recently introduced in Ref. 1 [B. Ph. van Milligen et al., Phys. Plasmas 11, 2272 (2004)], are studied. Different superdiffusive transport mechanisms generated by a family of distributions with algebraic decays (Tsallis distributions) are considered. It is observed that the possibility of changing the superdiffusive transport mechanism improves the flexibility of the model for describing different situations. The use of the model to describe the low (L) and high (H) confinement modes is also analyzed.

The Two-Dimensional Gabor Function Adapted to Natural Image Statistics: A Model of Simple-Cell Receptive Fields and Sparse Structure in Images.

PubMed

Loxley, P N

2017-10-01

The two-dimensional Gabor function is adapted to natural image statistics, leading to a tractable probabilistic generative model that can be used to model simple cell receptive field profiles, or generate basis functions for sparse coding applications. Learning is found to be most pronounced in three Gabor function parameters representing the size and spatial frequency of the two-dimensional Gabor function and characterized by a nonuniform probability distribution with heavy tails. All three parameters are found to be strongly correlated, resulting in a basis of multiscale Gabor functions with similar aspect ratios and size-dependent spatial frequencies. A key finding is that the distribution of receptive-field sizes is scale invariant over a wide range of values, so there is no characteristic receptive field size selected by natural image statistics. The Gabor function aspect ratio is found to be approximately conserved by the learning rules and is therefore not well determined by natural image statistics. This allows for three distinct solutions: a basis of Gabor functions with sharp orientation resolution at the expense of spatial-frequency resolution, a basis of Gabor functions with sharp spatial-frequency resolution at the expense of orientation resolution, or a basis with unit aspect ratio. Arbitrary mixtures of all three cases are also possible. Two parameters controlling the shape of the marginal distributions in a probabilistic generative model fully account for all three solutions. The best-performing probabilistic generative model for sparse coding applications is found to be a gaussian copula with Pareto marginal probability density functions.

Probabilistic Wind Power Ramp Forecasting Based on a Scenario Generation Method

DOE Office of Scientific and Technical Information (OSTI.GOV)

Wang, Qin; Florita, Anthony R; Krishnan, Venkat K

Wind power ramps (WPRs) are particularly important in the management and dispatch of wind power and currently drawing the attention of balancing authorities. With the aim to reduce the impact of WPRs for power system operations, this paper develops a probabilistic ramp forecasting method based on a large number of simulated scenarios. An ensemble machine learning technique is first adopted to forecast the basic wind power forecasting scenario and calculate the historical forecasting errors. A continuous Gaussian mixture model (GMM) is used to fit the probability distribution function (PDF) of forecasting errors. The cumulative distribution function (CDF) is analytically deduced.more » The inverse transform method based on Monte Carlo sampling and the CDF is used to generate a massive number of forecasting error scenarios. An optimized swinging door algorithm is adopted to extract all the WPRs from the complete set of wind power forecasting scenarios. The probabilistic forecasting results of ramp duration and start-time are generated based on all scenarios. Numerical simulations on publicly available wind power data show that within a predefined tolerance level, the developed probabilistic wind power ramp forecasting method is able to predict WPRs with a high level of sharpness and accuracy.« less

Probabilistic Wind Power Ramp Forecasting Based on a Scenario Generation Method: Preprint

DOE Office of Scientific and Technical Information (OSTI.GOV)

Wang, Qin; Florita, Anthony R; Krishnan, Venkat K

2017-08-31

Wind power ramps (WPRs) are particularly important in the management and dispatch of wind power, and they are currently drawing the attention of balancing authorities. With the aim to reduce the impact of WPRs for power system operations, this paper develops a probabilistic ramp forecasting method based on a large number of simulated scenarios. An ensemble machine learning technique is first adopted to forecast the basic wind power forecasting scenario and calculate the historical forecasting errors. A continuous Gaussian mixture model (GMM) is used to fit the probability distribution function (PDF) of forecasting errors. The cumulative distribution function (CDF) ismore » analytically deduced. The inverse transform method based on Monte Carlo sampling and the CDF is used to generate a massive number of forecasting error scenarios. An optimized swinging door algorithm is adopted to extract all the WPRs from the complete set of wind power forecasting scenarios. The probabilistic forecasting results of ramp duration and start time are generated based on all scenarios. Numerical simulations on publicly available wind power data show that within a predefined tolerance level, the developed probabilistic wind power ramp forecasting method is able to predict WPRs with a high level of sharpness and accuracy.« less

Maritime Threat Detection Using Probabilistic Graphical Models

DTIC Science & Technology

2012-01-01

CRF, unlike an HMM, can represent local features, and does not require feature concatenation. MLNs For MLNs, we used Alchemy ( Alchemy 2011), an...open source statistical relational learning and probabilistic inferencing package. Alchemy supports generative and discriminative weight learning, and...that Alchemy creates a new formula for every possible combination of the values for a1 and a2 that fit the type specified in their predicate

A probabilistic drought forecasting framework: A combined dynamical and statistical approach

DOE Office of Scientific and Technical Information (OSTI.GOV)

Yan, Hongxiang; Moradkhani, Hamid; Zarekarizi, Mahkameh

In order to improve drought forecasting skill, this study develops a probabilistic drought forecasting framework comprised of dynamical and statistical modeling components. The novelty of this study is to seek the use of data assimilation to quantify initial condition uncertainty with the Monte Carlo ensemble members, rather than relying entirely on the hydrologic model or land surface model to generate a single deterministic initial condition, as currently implemented in the operational drought forecasting systems. Next, the initial condition uncertainty is quantified through data assimilation and coupled with a newly developed probabilistic drought forecasting model using a copula function. The initialmore » condition at each forecast start date are sampled from the data assimilation ensembles for forecast initialization. Finally, seasonal drought forecasting products are generated with the updated initial conditions. This study introduces the theory behind the proposed drought forecasting system, with an application in Columbia River Basin, Pacific Northwest, United States. Results from both synthetic and real case studies suggest that the proposed drought forecasting system significantly improves the seasonal drought forecasting skills and can facilitate the state drought preparation and declaration, at least three months before the official state drought declaration.« less

NEXT GENERATION MULTIMEDIA/MULTIPATHWAY EXPOSURE MODELING

EPA Science Inventory

The Stochastic Human Exposure and Dose Simulation model for pesticides (SHEDS-Pesticides) supports the efforts of EPA to better understand human exposures and doses to multimedia, multipathway pollutants. It is a physically-based, probabilistic computer model that predicts, for u...

Adequacy assessment of composite generation and transmission systems incorporating wind energy conversion systems

NASA Astrophysics Data System (ADS)

Gao, Yi

The development and utilization of wind energy for satisfying electrical demand has received considerable attention in recent years due to its tremendous environmental, social and economic benefits, together with public support and government incentives. Electric power generation from wind energy behaves quite differently from that of conventional sources. The fundamentally different operating characteristics of wind energy facilities therefore affect power system reliability in a different manner than those of conventional systems. The reliability impact of such a highly variable energy source is an important aspect that must be assessed when the wind power penetration is significant. The focus of the research described in this thesis is on the utilization of state sampling Monte Carlo simulation in wind integrated bulk electric system reliability analysis and the application of these concepts in system planning and decision making. Load forecast uncertainty is an important factor in long range planning and system development. This thesis describes two approximate approaches developed to reduce the number of steps in a load duration curve which includes load forecast uncertainty, and to provide reasonably accurate generating and bulk system reliability index predictions. The developed approaches are illustrated by application to two composite test systems. A method of generating correlated random numbers with uniform distributions and a specified correlation coefficient in the state sampling method is proposed and used to conduct adequacy assessment in generating systems and in bulk electric systems containing correlated wind farms in this thesis. The studies described show that it is possible to use the state sampling Monte Carlo simulation technique to quantitatively assess the reliability implications associated with adding wind power to a composite generation and transmission system including the effects of multiple correlated wind sites. This is an important development as it permits correlated wind farms to be incorporated in large practical system studies without requiring excessive increases in computer solution time. The procedures described in this thesis for creating monthly and seasonal wind farm models should prove useful in situations where time period models are required to incorporate scheduled maintenance of generation and transmission facilities. There is growing interest in combining deterministic considerations with probabilistic assessment in order to evaluate the quantitative system risk and conduct bulk power system planning. A relatively new approach that incorporates deterministic and probabilistic considerations in a single risk assessment framework has been designated as the joint deterministic-probabilistic approach. The research work described in this thesis illustrates that the joint deterministic-probabilistic approach can be effectively used to integrate wind power in bulk electric system planning. The studies described in this thesis show that the application of the joint deterministic-probabilistic method provides more stringent results for a system with wind power than the traditional deterministic N-1 method because the joint deterministic-probabilistic technique is driven by the deterministic N-1 criterion with an added probabilistic perspective which recognizes the power output characteristics of a wind turbine generator.

Comparison of Control Approaches in Genetic Regulatory Networks by Using Stochastic Master Equation Models, Probabilistic Boolean Network Models and Differential Equation Models and Estimated Error Analyzes

NASA Astrophysics Data System (ADS)

Caglar, Mehmet Umut; Pal, Ranadip

2011-03-01

Central dogma of molecular biology states that ``information cannot be transferred back from protein to either protein or nucleic acid''. However, this assumption is not exactly correct in most of the cases. There are a lot of feedback loops and interactions between different levels of systems. These types of interactions are hard to analyze due to the lack of cell level data and probabilistic - nonlinear nature of interactions. Several models widely used to analyze and simulate these types of nonlinear interactions. Stochastic Master Equation (SME) models give probabilistic nature of the interactions in a detailed manner, with a high calculation cost. On the other hand Probabilistic Boolean Network (PBN) models give a coarse scale picture of the stochastic processes, with a less calculation cost. Differential Equation (DE) models give the time evolution of mean values of processes in a highly cost effective way. The understanding of the relations between the predictions of these models is important to understand the reliability of the simulations of genetic regulatory networks. In this work the success of the mapping between SME, PBN and DE models is analyzed and the accuracy and affectivity of the control policies generated by using PBN and DE models is compared.

Generating short-term probabilistic wind power scenarios via nonparametric forecast error density estimators: Generating short-term probabilistic wind power scenarios via nonparametric forecast error density estimators

DOE PAGES

Staid, Andrea; Watson, Jean -Paul; Wets, Roger J. -B.; ...

2017-07-11

Forecasts of available wind power are critical in key electric power systems operations planning problems, including economic dispatch and unit commitment. Such forecasts are necessarily uncertain, limiting the reliability and cost effectiveness of operations planning models based on a single deterministic or “point” forecast. A common approach to address this limitation involves the use of a number of probabilistic scenarios, each specifying a possible trajectory of wind power production, with associated probability. We present and analyze a novel method for generating probabilistic wind power scenarios, leveraging available historical information in the form of forecasted and corresponding observed wind power timemore » series. We estimate non-parametric forecast error densities, specifically using epi-spline basis functions, allowing us to capture the skewed and non-parametric nature of error densities observed in real-world data. We then describe a method to generate probabilistic scenarios from these basis functions that allows users to control for the degree to which extreme errors are captured.We compare the performance of our approach to the current state-of-the-art considering publicly available data associated with the Bonneville Power Administration, analyzing aggregate production of a number of wind farms over a large geographic region. Finally, we discuss the advantages of our approach in the context of specific power systems operations planning problems: stochastic unit commitment and economic dispatch. Here, our methodology is embodied in the joint Sandia – University of California Davis Prescient software package for assessing and analyzing stochastic operations strategies.« less

Generating short-term probabilistic wind power scenarios via nonparametric forecast error density estimators: Generating short-term probabilistic wind power scenarios via nonparametric forecast error density estimators

DOE Office of Scientific and Technical Information (OSTI.GOV)

Staid, Andrea; Watson, Jean -Paul; Wets, Roger J. -B.

Forecasts of available wind power are critical in key electric power systems operations planning problems, including economic dispatch and unit commitment. Such forecasts are necessarily uncertain, limiting the reliability and cost effectiveness of operations planning models based on a single deterministic or “point” forecast. A common approach to address this limitation involves the use of a number of probabilistic scenarios, each specifying a possible trajectory of wind power production, with associated probability. We present and analyze a novel method for generating probabilistic wind power scenarios, leveraging available historical information in the form of forecasted and corresponding observed wind power timemore » series. We estimate non-parametric forecast error densities, specifically using epi-spline basis functions, allowing us to capture the skewed and non-parametric nature of error densities observed in real-world data. We then describe a method to generate probabilistic scenarios from these basis functions that allows users to control for the degree to which extreme errors are captured.We compare the performance of our approach to the current state-of-the-art considering publicly available data associated with the Bonneville Power Administration, analyzing aggregate production of a number of wind farms over a large geographic region. Finally, we discuss the advantages of our approach in the context of specific power systems operations planning problems: stochastic unit commitment and economic dispatch. Here, our methodology is embodied in the joint Sandia – University of California Davis Prescient software package for assessing and analyzing stochastic operations strategies.« less

MODELING ENVIRONMENTAL EXPOSURES TO PARTICULATE MATTER AND PESTICIDES

EPA Science Inventory

This presentation describes initial results from on-going research at EPA on modeling human exposures to particulate matter and residential pesticides. A first generation probabilistic population exposure model for Particulate Matter (PM), specifically for predicting PM1o and P...

The Epistemic Representation of Information Flow Security in Probabilistic Systems

DTIC Science & Technology

1995-06-01

The new characterization also means that our security crite- rion is expressible in a simpler logic and model. 1 Introduction Multilevel security is...ber generator) during its execution. Such probabilistic choices are useful in a multilevel security context for Supported by grants HKUST 608/94E from... 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and

Technical report. The application of probability-generating functions to linear-quadratic radiation survival curves.

PubMed

Kendal, W S

2000-04-01

To illustrate how probability-generating functions (PGFs) can be employed to derive a simple probabilistic model for clonogenic survival after exposure to ionizing irradiation. Both repairable and irreparable radiation damage to DNA were assumed to occur by independent (Poisson) processes, at intensities proportional to the irradiation dose. Also, repairable damage was assumed to be either repaired or further (lethally) injured according to a third (Bernoulli) process, with the probability of lethal conversion being directly proportional to dose. Using the algebra of PGFs, these three processes were combined to yield a composite PGF that described the distribution of lethal DNA lesions in irradiated cells. The composite PGF characterized a Poisson distribution with mean, chiD+betaD2, where D was dose and alpha and beta were radiobiological constants. This distribution yielded the conventional linear-quadratic survival equation. To test the composite model, the derived distribution was used to predict the frequencies of multiple chromosomal aberrations in irradiated human lymphocytes. The predictions agreed well with observation. This probabilistic model was consistent with single-hit mechanisms, but it was not consistent with binary misrepair mechanisms. A stochastic model for radiation survival has been constructed from elementary PGFs that exactly yields the linear-quadratic relationship. This approach can be used to investigate other simple probabilistic survival models.

Probabilistic topic modeling for the analysis and classification of genomic sequences

PubMed Central

2015-01-01

Background Studies on genomic sequences for classification and taxonomic identification have a leading role in the biomedical field and in the analysis of biodiversity. These studies are focusing on the so-called barcode genes, representing a well defined region of the whole genome. Recently, alignment-free techniques are gaining more importance because they are able to overcome the drawbacks of sequence alignment techniques. In this paper a new alignment-free method for DNA sequences clustering and classification is proposed. The method is based on k-mers representation and text mining techniques. Methods The presented method is based on Probabilistic Topic Modeling, a statistical technique originally proposed for text documents. Probabilistic topic models are able to find in a document corpus the topics (recurrent themes) characterizing classes of documents. This technique, applied on DNA sequences representing the documents, exploits the frequency of fixed-length k-mers and builds a generative model for a training group of sequences. This generative model, obtained through the Latent Dirichlet Allocation (LDA) algorithm, is then used to classify a large set of genomic sequences. Results and conclusions We performed classification of over 7000 16S DNA barcode sequences taken from Ribosomal Database Project (RDP) repository, training probabilistic topic models. The proposed method is compared to the RDP tool and Support Vector Machine (SVM) classification algorithm in a extensive set of trials using both complete sequences and short sequence snippets (from 400 bp to 25 bp). Our method reaches very similar results to RDP classifier and SVM for complete sequences. The most interesting results are obtained when short sequence snippets are considered. In these conditions the proposed method outperforms RDP and SVM with ultra short sequences and it exhibits a smooth decrease of performance, at every taxonomic level, when the sequence length is decreased. PMID:25916734

Incorporating seismic phase correlations into a probabilistic model of global-scale seismology

NASA Astrophysics Data System (ADS)

Arora, Nimar

2013-04-01

We present a probabilistic model of seismic phases whereby the attributes of the body-wave phases are correlated to those of the first arriving P phase. This model has been incorporated into NET-VISA (Network processing Vertically Integrated Seismic Analysis) a probabilistic generative model of seismic events, their transmission, and detection on a global seismic network. In the earlier version of NET-VISA, seismic phase were assumed to be independent of each other. Although this didn't affect the quality of the inferred seismic bulletin, for the most part, it did result in a few instances of anomalous phase association. For example, an S phase with a smaller slowness than the corresponding P phase. We demonstrate that the phase attributes are indeed highly correlated, for example the uncertainty in the S phase travel time is significantly reduced given the P phase travel time. Our new model exploits these correlations to produce better calibrated probabilities for the events, as well as fewer anomalous associations.

Probabilistic sensitivity analysis for decision trees with multiple branches: use of the Dirichlet distribution in a Bayesian framework.

PubMed

Briggs, Andrew H; Ades, A E; Price, Martin J

2003-01-01

In structuring decision models of medical interventions, it is commonly recommended that only 2 branches be used for each chance node to avoid logical inconsistencies that can arise during sensitivity analyses if the branching probabilities do not sum to 1. However, information may be naturally available in an unconditional form, and structuring a tree in conditional form may complicate rather than simplify the sensitivity analysis of the unconditional probabilities. Current guidance emphasizes using probabilistic sensitivity analysis, and a method is required to provide probabilistic probabilities over multiple branches that appropriately represents uncertainty while satisfying the requirement that mutually exclusive event probabilities should sum to 1. The authors argue that the Dirichlet distribution, the multivariate equivalent of the beta distribution, is appropriate for this purpose and illustrate its use for generating a fully probabilistic transition matrix for a Markov model. Furthermore, they demonstrate that by adopting a Bayesian approach, the problem of observing zero counts for transitions of interest can be overcome.

Probabilistic simple sticker systems

NASA Astrophysics Data System (ADS)

Selvarajoo, Mathuri; Heng, Fong Wan; Sarmin, Nor Haniza; Turaev, Sherzod

2017-04-01

A model for DNA computing using the recombination behavior of DNA molecules, known as a sticker system, was introduced by by L. Kari, G. Paun, G. Rozenberg, A. Salomaa, and S. Yu in the paper entitled DNA computing, sticker systems and universality from the journal of Acta Informatica vol. 35, pp. 401-420 in the year 1998. A sticker system uses the Watson-Crick complementary feature of DNA molecules: starting from the incomplete double stranded sequences, and iteratively using sticking operations until a complete double stranded sequence is obtained. It is known that sticker systems with finite sets of axioms and sticker rules generate only regular languages. Hence, different types of restrictions have been considered to increase the computational power of sticker systems. Recently, a variant of restricted sticker systems, called probabilistic sticker systems, has been introduced [4]. In this variant, the probabilities are initially associated with the axioms, and the probability of a generated string is computed by multiplying the probabilities of all occurrences of the initial strings in the computation of the string. Strings for the language are selected according to some probabilistic requirements. In this paper, we study fundamental properties of probabilistic simple sticker systems. We prove that the probabilistic enhancement increases the computational power of simple sticker systems.

A probabilistic seismic model for the European Arctic

NASA Astrophysics Data System (ADS)

Hauser, Juerg; Dyer, Kathleen M.; Pasyanos, Michael E.; Bungum, Hilmar; Faleide, Jan I.; Clark, Stephen A.; Schweitzer, Johannes

2011-01-01

The development of three-dimensional seismic models for the crust and upper mantle has traditionally focused on finding one model that provides the best fit to the data while observing some regularization constraints. In contrast to this, the inversion employed here fits the data in a probabilistic sense and thus provides a quantitative measure of model uncertainty. Our probabilistic model is based on two sources of information: (1) prior information, which is independent from the data, and (2) different geophysical data sets, including thickness constraints, velocity profiles, gravity data, surface wave group velocities, and regional body wave traveltimes. We use a Markov chain Monte Carlo (MCMC) algorithm to sample models from the prior distribution, the set of plausible models, and test them against the data to generate the posterior distribution, the ensemble of models that fit the data with assigned uncertainties. While being computationally more expensive, such a probabilistic inversion provides a more complete picture of solution space and allows us to combine various data sets. The complex geology of the European Arctic, encompassing oceanic crust, continental shelf regions, rift basins and old cratonic crust, as well as the nonuniform coverage of the region by data with varying degrees of uncertainty, makes it a challenging setting for any imaging technique and, therefore, an ideal environment for demonstrating the practical advantages of a probabilistic approach. Maps of depth to basement and depth to Moho derived from the posterior distribution are in good agreement with previously published maps and interpretations of the regional tectonic setting. The predicted uncertainties, which are as important as the absolute values, correlate well with the variations in data coverage and quality in the region. A practical advantage of our probabilistic model is that it can provide estimates for the uncertainties of observables due to model uncertainties. We will demonstrate how this can be used for the formulation of earthquake location algorithms that take model uncertainties into account when estimating location uncertainties.

Price of Fairness in Kidney Exchange

DTIC Science & Technology

2014-05-01

solver uses branch-and-price, a technique that proves optimality by in- crementally generating only a small part of the model during tree search [8...factors like fail- ure probability and chain position, as in the probabilistic model ). We will use this multiplicative re-weighting in our experiments in...Table 2 gives the average loss in efficiency for each of these models over multiple generated pool sizes, with 40 runs per pool size per model , under

A Probabilistic Model of Illegal Drug Trafficking Operations in the Eastern Pacific and Caribbean Sea

DTIC Science & Technology

2013-09-01

partner agencies and nations, detects, tracks, and interdicts illegal drug-trafficking in this region. In this thesis, we develop a probability model based...trafficking in this region. In this thesis, we develop a probability model based on intelligence inputs to generate a spatial temporal heat map specifying the...complement and vet such complicated simulation by developing more analytically tractable models. We develop probability models to generate a heat map

Automation on the generation of genome-scale metabolic models.

PubMed

Reyes, R; Gamermann, D; Montagud, A; Fuente, D; Triana, J; Urchueguía, J F; de Córdoba, P Fernández

2012-12-01

Nowadays, the reconstruction of genome-scale metabolic models is a nonautomatized and interactive process based on decision making. This lengthy process usually requires a full year of one person's work in order to satisfactory collect, analyze, and validate the list of all metabolic reactions present in a specific organism. In order to write this list, one manually has to go through a huge amount of genomic, metabolomic, and physiological information. Currently, there is no optimal algorithm that allows one to automatically go through all this information and generate the models taking into account probabilistic criteria of unicity and completeness that a biologist would consider. This work presents the automation of a methodology for the reconstruction of genome-scale metabolic models for any organism. The methodology that follows is the automatized version of the steps implemented manually for the reconstruction of the genome-scale metabolic model of a photosynthetic organism, Synechocystis sp. PCC6803. The steps for the reconstruction are implemented in a computational platform (COPABI) that generates the models from the probabilistic algorithms that have been developed. For validation of the developed algorithm robustness, the metabolic models of several organisms generated by the platform have been studied together with published models that have been manually curated. Network properties of the models, like connectivity and average shortest mean path of the different models, have been compared and analyzed.

Probabilistic #D data fusion for multiresolution surface generation

NASA Technical Reports Server (NTRS)

Manduchi, R.; Johnson, A. E.

2002-01-01

In this paper we present an algorithm for adaptive resolution integration of 3D data collected from multiple distributed sensors. The input to the algorithm is a set of 3D surface points and associated sensor models. Using a probabilistic rule, a surface probability function is generated that represents the probability that a particular volume of space contains the surface. The surface probability function is represented using an octree data structure; regions of space with samples of large conariance are stored at a coarser level than regions of space containing samples with smaller covariance. The algorithm outputs an adaptive resolution surface generated by connecting points that lie on the ridge of surface probability with triangles scaled to match the local discretization of space given by the algorithm, we present results from 3D data generated by scanning lidar and structure from motion.

ORNL Pre-test Analyses of A Large-scale Experiment in STYLE

DOE Office of Scientific and Technical Information (OSTI.GOV)

Williams, Paul T; Yin, Shengjun; Klasky, Hilda B

Oak Ridge National Laboratory (ORNL) is conducting a series of numerical analyses to simulate a large scale mock-up experiment planned within the European Network for Structural Integrity for Lifetime Management non-RPV Components (STYLE). STYLE is a European cooperative effort to assess the structural integrity of (non-reactor pressure vessel) reactor coolant pressure boundary components relevant to ageing and life-time management and to integrate the knowledge created in the project into mainstream nuclear industry assessment codes. ORNL contributes work-in-kind support to STYLE Work Package 2 (Numerical Analysis/Advanced Tools) and Work Package 3 (Engineering Assessment Methods/LBB Analyses). This paper summarizes the current statusmore » of ORNL analyses of the STYLE Mock-Up3 large-scale experiment to simulate and evaluate crack growth in a cladded ferritic pipe. The analyses are being performed in two parts. In the first part, advanced fracture mechanics models are being developed and performed to evaluate several experiment designs taking into account the capabilities of the test facility while satisfying the test objectives. Then these advanced fracture mechanics models will be utilized to simulate the crack growth in the large scale mock-up test. For the second part, the recently developed ORNL SIAM-PFM open-source, cross-platform, probabilistic computational tool will be used to generate an alternative assessment for comparison with the advanced fracture mechanics model results. The SIAM-PFM probabilistic analysis of the Mock-Up3 experiment will utilize fracture modules that are installed into a general probabilistic framework. The probabilistic results of the Mock-Up3 experiment obtained from SIAM-PFM will be compared to those results generated using the deterministic 3D nonlinear finite-element modeling approach. The objective of the probabilistic analysis is to provide uncertainty bounds that will assist in assessing the more detailed 3D finite-element solutions and to also assess the level of confidence that can be placed in the best-estimate finiteelement solutions.« less

Learning Orthographic Structure with Sequential Generative Neural Networks

ERIC Educational Resources Information Center

Testolin, Alberto; Stoianov, Ivilin; Sperduti, Alessandro; Zorzi, Marco

2016-01-01

Learning the structure of event sequences is a ubiquitous problem in cognition and particularly in language. One possible solution is to learn a probabilistic generative model of sequences that allows making predictions about upcoming events. Though appealing from a neurobiological standpoint, this approach is typically not pursued in…

Probabilistic analysis for fatigue strength degradation of materials

NASA Technical Reports Server (NTRS)

Royce, Lola

1989-01-01

This report presents the results of the first year of a research program conducted for NASA-LeRC by the University of Texas at San Antonio. The research included development of methodology that provides a probabilistic treatment of lifetime prediction of structural components of aerospace propulsion systems subjected to fatigue. Material strength degradation models, based on primitive variables, include both a fatigue strength reduction model and a fatigue crack growth model. Linear elastic fracture mechanics is utilized in the latter model. Probabilistic analysis is based on simulation, and both maximum entropy and maximum penalized likelihood methods are used for the generation of probability density functions. The resulting constitutive relationships are included in several computer programs, RANDOM2, RANDOM3, and RANDOM4. These programs determine the random lifetime of an engine component, in mechanical load cycles, to reach a critical fatigue strength or crack size. The material considered was a cast nickel base superalloy, one typical of those used in the Space Shuttle Main Engine.

Synthetic wind speed scenarios generation for probabilistic analysis of hybrid energy systems

DOE Office of Scientific and Technical Information (OSTI.GOV)

Chen, Jun; Rabiti, Cristian

Hybrid energy systems consisting of multiple energy inputs and multiple energy outputs have been proposed to be an effective element to enable ever increasing penetration of clean energy. In order to better understand the dynamic and probabilistic behavior of hybrid energy systems, this paper proposes a model combining Fourier series and autoregressive moving average (ARMA) to characterize historical weather measurements and to generate synthetic weather (e.g., wind speed) data. In particular, Fourier series is used to characterize the seasonal trend in historical data, while ARMA is applied to capture the autocorrelation in residue time series (e.g., measurements minus seasonal trends).more » The generated synthetic wind speed data is then utilized to perform probabilistic analysis of a particular hybrid energy system con guration, which consists of nuclear power plant, wind farm, battery storage, natural gas boiler, and chemical plant. As a result, requirements on component ramping rate, economic and environmental impacts of hybrid energy systems, and the effects of deploying different sizes of batteries in smoothing renewable variability, are all investigated.« less

Synthetic wind speed scenarios generation for probabilistic analysis of hybrid energy systems

DOE PAGES

Chen, Jun; Rabiti, Cristian

2016-11-25

Hybrid energy systems consisting of multiple energy inputs and multiple energy outputs have been proposed to be an effective element to enable ever increasing penetration of clean energy. In order to better understand the dynamic and probabilistic behavior of hybrid energy systems, this paper proposes a model combining Fourier series and autoregressive moving average (ARMA) to characterize historical weather measurements and to generate synthetic weather (e.g., wind speed) data. In particular, Fourier series is used to characterize the seasonal trend in historical data, while ARMA is applied to capture the autocorrelation in residue time series (e.g., measurements minus seasonal trends).more » The generated synthetic wind speed data is then utilized to perform probabilistic analysis of a particular hybrid energy system con guration, which consists of nuclear power plant, wind farm, battery storage, natural gas boiler, and chemical plant. As a result, requirements on component ramping rate, economic and environmental impacts of hybrid energy systems, and the effects of deploying different sizes of batteries in smoothing renewable variability, are all investigated.« less

Probabilistic structural analysis of space propulsion system LOX post

NASA Technical Reports Server (NTRS)

Newell, J. F.; Rajagopal, K. R.; Ho, H. W.; Cunniff, J. M.

1990-01-01

The probabilistic structural analysis program NESSUS (Numerical Evaluation of Stochastic Structures Under Stress; Cruse et al., 1988) is applied to characterize the dynamic loading and response of the Space Shuttle main engine (SSME) LOX post. The design and operation of the SSME are reviewed; the LOX post structure is described; and particular attention is given to the generation of composite load spectra, the finite-element model of the LOX post, and the steps in the NESSUS structural analysis. The results are presented in extensive tables and graphs, and it is shown that NESSUS correctly predicts the structural effects of changes in the temperature loading. The probabilistic approach also facilitates (1) damage assessments for a given failure model (based on gas temperature, heat-shield gap, and material properties) and (2) correlation of the gas temperature with operational parameters such as engine thrust.

INTEGRATED PROBABILISTIC AND DETERMINISTIC MODELING TECHNIQUES IN ESTIMATING EXPOSURE TO WATER-BORNE CONTAMINANTS: PART 2 PHARMACOKINETIC MODELING

EPA Science Inventory

The Total Exposure Model (TEM) uses deterministic and stochastic methods to estimate the exposure of a person performing daily activities of eating, drinking, showering, and bathing. There were 250 time histories generated, by subject with activities, for the three exposure ro...

DOE Office of Scientific and Technical Information (OSTI.GOV)

Ng, B

This survey gives an overview of popular generative models used in the modeling of stochastic temporal systems. In particular, this survey is organized into two parts. The first part discusses the discrete-time representations of dynamic Bayesian networks and dynamic relational probabilistic models, while the second part discusses the continuous-time representation of continuous-time Bayesian networks.

Analysis of flood hazard under consideration of dike breaches

NASA Astrophysics Data System (ADS)

Vorogushyn, S.; Apel, H.; Lindenschmidt, K.-E.; Merz, B.

2009-04-01

The study focuses on the development and application of a new modelling system which allows a comprehensive flood hazard assessment along diked river reaches under consideration of dike failures. The proposed Inundation Hazard Assessment Model (IHAM) represents a hybrid probabilistic-deterministic model. It comprises three models interactively coupled at runtime. These are: (1) 1D unsteady hydrodynamic model of river channel and floodplain flow between dikes, (2) probabilistic dike breach model which determines possible dike breach locations, breach widths and breach outflow discharges, and (3) 2D raster-based diffusion wave storage cell model of the hinterland areas behind the dikes. Due to the unsteady nature of the 1D and 2D coupled models, the dependence between hydraulic load at various locations along the reach is explicitly considered. The probabilistic dike breach model describes dike failures due to three failure mechanisms: overtopping, piping and slope instability caused by the seepage flow through the dike core (micro-instability). Dike failures for each mechanism are simulated based on fragility functions. The probability of breach is conditioned by the uncertainty in geometrical and geotechnical dike parameters. The 2D storage cell model driven by the breach outflow boundary conditions computes an extended spectrum of flood intensity indicators such as water depth, flow velocity, impulse, inundation duration and rate of water rise. IHAM is embedded in a Monte Carlo simulation in order to account for the natural variability of the flood generation processes reflected in the form of input hydrographs and for the randomness of dike failures given by breach locations, times and widths. The scenario calculations for the developed synthetic input hydrographs for the main river and tributary were carried out for floods with return periods of T = 100; 200; 500; 1000 a. Based on the modelling results, probabilistic dike hazard maps could be generated that indicate the failure probability of each discretised dike section for every scenario magnitude. Besides the binary inundation patterns that indicate the probability of raster cells being inundated, IHAM generates probabilistic flood hazard maps. These maps display spatial patterns of the considered flood intensity indicators and their associated return periods. The probabilistic nature of IHAM allows for the generation of percentile flood hazard maps that indicate the median and uncertainty bounds of the flood intensity indicators. The uncertainty results from the natural variability of the flow hydrographs and randomness of dike breach processes. The same uncertainty sources determine the uncertainty in the flow hydrographs along the study reach. The simulations showed that the dike breach stochasticity has an increasing impact on hydrograph uncertainty in downstream direction. Whereas in the upstream part of the reach the hydrograph uncertainty is mainly stipulated by the variability of the flood wave form, the dike failures strongly shape the uncertainty boundaries in the downstream part of the reach. Finally, scenarios of polder deployment for the extreme floods with T = 200; 500; 1000 a were simulated with IHAM. The results indicate a rather weak reduction of the mean and median flow hydrographs in the river channel. However, the capping of the flow peaks resulted in a considerable reduction of the overtopping failures downstream of the polder with a simultaneous slight increase of the piping and slope micro-instability frequencies explained by a more durable average impoundment. The developed IHAM simulation system represents a new scientific tool for studying fluvial inundation dynamics under extreme conditions incorporating effects of technical flood protection measures. With its major outputs in form of novel probabilistic inundation and dike hazard maps, the IHAM system has a high practical value for decision support in flood management.

Predicting inpatient clinical order patterns with probabilistic topic models vs conventional order sets.

PubMed

Chen, Jonathan H; Goldstein, Mary K; Asch, Steven M; Mackey, Lester; Altman, Russ B

2017-05-01

Build probabilistic topic model representations of hospital admissions processes and compare the ability of such models to predict clinical order patterns as compared to preconstructed order sets. The authors evaluated the first 24 hours of structured electronic health record data for > 10 K inpatients. Drawing an analogy between structured items (e.g., clinical orders) to words in a text document, the authors performed latent Dirichlet allocation probabilistic topic modeling. These topic models use initial clinical information to predict clinical orders for a separate validation set of > 4 K patients. The authors evaluated these topic model-based predictions vs existing human-authored order sets by area under the receiver operating characteristic curve, precision, and recall for subsequent clinical orders. Existing order sets predict clinical orders used within 24 hours with area under the receiver operating characteristic curve 0.81, precision 16%, and recall 35%. This can be improved to 0.90, 24%, and 47% ( P  < 10 -20 ) by using probabilistic topic models to summarize clinical data into up to 32 topics. Many of these latent topics yield natural clinical interpretations (e.g., "critical care," "pneumonia," "neurologic evaluation"). Existing order sets tend to provide nonspecific, process-oriented aid, with usability limitations impairing more precise, patient-focused support. Algorithmic summarization has the potential to breach this usability barrier by automatically inferring patient context, but with potential tradeoffs in interpretability. Probabilistic topic modeling provides an automated approach to detect thematic trends in patient care and generate decision support content. A potential use case finds related clinical orders for decision support. © The Author 2016. Published by Oxford University Press on behalf of the American Medical Informatics Association.

Predicting inpatient clinical order patterns with probabilistic topic models vs conventional order sets

PubMed Central

Goldstein, Mary K; Asch, Steven M; Mackey, Lester; Altman, Russ B

2017-01-01

Objective: Build probabilistic topic model representations of hospital admissions processes and compare the ability of such models to predict clinical order patterns as compared to preconstructed order sets. Materials and Methods: The authors evaluated the first 24 hours of structured electronic health record data for > 10 K inpatients. Drawing an analogy between structured items (e.g., clinical orders) to words in a text document, the authors performed latent Dirichlet allocation probabilistic topic modeling. These topic models use initial clinical information to predict clinical orders for a separate validation set of > 4 K patients. The authors evaluated these topic model-based predictions vs existing human-authored order sets by area under the receiver operating characteristic curve, precision, and recall for subsequent clinical orders. Results: Existing order sets predict clinical orders used within 24 hours with area under the receiver operating characteristic curve 0.81, precision 16%, and recall 35%. This can be improved to 0.90, 24%, and 47% (P < 10−20) by using probabilistic topic models to summarize clinical data into up to 32 topics. Many of these latent topics yield natural clinical interpretations (e.g., “critical care,” “pneumonia,” “neurologic evaluation”). Discussion: Existing order sets tend to provide nonspecific, process-oriented aid, with usability limitations impairing more precise, patient-focused support. Algorithmic summarization has the potential to breach this usability barrier by automatically inferring patient context, but with potential tradeoffs in interpretability. Conclusion: Probabilistic topic modeling provides an automated approach to detect thematic trends in patient care and generate decision support content. A potential use case finds related clinical orders for decision support. PMID:27655861

Data Analysis with Graphical Models: Software Tools

NASA Technical Reports Server (NTRS)

Buntine, Wray L.

1994-01-01

Probabilistic graphical models (directed and undirected Markov fields, and combined in chain graphs) are used widely in expert systems, image processing and other areas as a framework for representing and reasoning with probabilities. They come with corresponding algorithms for performing probabilistic inference. This paper discusses an extension to these models by Spiegelhalter and Gilks, plates, used to graphically model the notion of a sample. This offers a graphical specification language for representing data analysis problems. When combined with general methods for statistical inference, this also offers a unifying framework for prototyping and/or generating data analysis algorithms from graphical specifications. This paper outlines the framework and then presents some basic tools for the task: a graphical version of the Pitman-Koopman Theorem for the exponential family, problem decomposition, and the calculation of exact Bayes factors. Other tools already developed, such as automatic differentiation, Gibbs sampling, and use of the EM algorithm, make this a broad basis for the generation of data analysis software.

Probabilistic tsunami hazard analysis: Multiple sources and global applications

USGS Publications Warehouse

Grezio, Anita; Babeyko, Andrey; Baptista, Maria Ana; Behrens, Jörn; Costa, Antonio; Davies, Gareth; Geist, Eric L.; Glimsdal, Sylfest; González, Frank I.; Griffin, Jonathan; Harbitz, Carl B.; LeVeque, Randall J.; Lorito, Stefano; Løvholt, Finn; Omira, Rachid; Mueller, Christof; Paris, Raphaël; Parsons, Thomas E.; Polet, Jascha; Power, William; Selva, Jacopo; Sørensen, Mathilde B.; Thio, Hong Kie

2017-01-01

Applying probabilistic methods to infrequent but devastating natural events is intrinsically challenging. For tsunami analyses, a suite of geophysical assessments should be in principle evaluated because of the different causes generating tsunamis (earthquakes, landslides, volcanic activity, meteorological events, and asteroid impacts) with varying mean recurrence rates. Probabilistic Tsunami Hazard Analyses (PTHAs) are conducted in different areas of the world at global, regional, and local scales with the aim of understanding tsunami hazard to inform tsunami risk reduction activities. PTHAs enhance knowledge of the potential tsunamigenic threat by estimating the probability of exceeding specific levels of tsunami intensity metrics (e.g., run-up or maximum inundation heights) within a certain period of time (exposure time) at given locations (target sites); these estimates can be summarized in hazard maps or hazard curves. This discussion presents a broad overview of PTHA, including (i) sources and mechanisms of tsunami generation, emphasizing the variety and complexity of the tsunami sources and their generation mechanisms, (ii) developments in modeling the propagation and impact of tsunami waves, and (iii) statistical procedures for tsunami hazard estimates that include the associated epistemic and aleatoric uncertainties. Key elements in understanding the potential tsunami hazard are discussed, in light of the rapid development of PTHA methods during the last decade and the globally distributed applications, including the importance of considering multiple sources, their relative intensities, probabilities of occurrence, and uncertainties in an integrated and consistent probabilistic framework.

Probabilistic Tsunami Hazard Analysis: Multiple Sources and Global Applications

NASA Astrophysics Data System (ADS)

Grezio, Anita; Babeyko, Andrey; Baptista, Maria Ana; Behrens, Jörn; Costa, Antonio; Davies, Gareth; Geist, Eric L.; Glimsdal, Sylfest; González, Frank I.; Griffin, Jonathan; Harbitz, Carl B.; LeVeque, Randall J.; Lorito, Stefano; Løvholt, Finn; Omira, Rachid; Mueller, Christof; Paris, Raphaël.; Parsons, Tom; Polet, Jascha; Power, William; Selva, Jacopo; Sørensen, Mathilde B.; Thio, Hong Kie

2017-12-01

Applying probabilistic methods to infrequent but devastating natural events is intrinsically challenging. For tsunami analyses, a suite of geophysical assessments should be in principle evaluated because of the different causes generating tsunamis (earthquakes, landslides, volcanic activity, meteorological events, and asteroid impacts) with varying mean recurrence rates. Probabilistic Tsunami Hazard Analyses (PTHAs) are conducted in different areas of the world at global, regional, and local scales with the aim of understanding tsunami hazard to inform tsunami risk reduction activities. PTHAs enhance knowledge of the potential tsunamigenic threat by estimating the probability of exceeding specific levels of tsunami intensity metrics (e.g., run-up or maximum inundation heights) within a certain period of time (exposure time) at given locations (target sites); these estimates can be summarized in hazard maps or hazard curves. This discussion presents a broad overview of PTHA, including (i) sources and mechanisms of tsunami generation, emphasizing the variety and complexity of the tsunami sources and their generation mechanisms, (ii) developments in modeling the propagation and impact of tsunami waves, and (iii) statistical procedures for tsunami hazard estimates that include the associated epistemic and aleatoric uncertainties. Key elements in understanding the potential tsunami hazard are discussed, in light of the rapid development of PTHA methods during the last decade and the globally distributed applications, including the importance of considering multiple sources, their relative intensities, probabilities of occurrence, and uncertainties in an integrated and consistent probabilistic framework.

High-Resolution Underwater Mapping Using Side-Scan Sonar

PubMed Central

2016-01-01

The goal of this study is to generate high-resolution sea floor maps using a Side-Scan Sonar(SSS). This is achieved by explicitly taking into account the SSS operation as follows. First, the raw sensor data is corrected by means of a physics-based SSS model. Second, the data is projected to the sea-floor. The errors involved in this projection are thoroughfully analysed. Third, a probabilistic SSS model is defined and used to estimate the probability of each sea-floor region to be observed. This probabilistic information is then used to weight the contribution of each SSS measurement to the map. Because of these models, arbitrary map resolutions can be achieved, even beyond the sensor resolution. Finally, a geometric map building method is presented and combined with the probabilistic approach. The resulting map is composed of two layers. The echo intensity layer holds the most likely echo intensities at each point in the sea-floor. The probabilistic layer contains information about how confident can the user or the higher control layers be about the echo intensity layer data. Experimental results have been conducted in a large subsea region. PMID:26821379

What is the Value Added to Adaptation Planning by Probabilistic Projections of Climate Change?

NASA Astrophysics Data System (ADS)

Wilby, R. L.

2008-12-01

Probabilistic projections of climate change offer new sources of risk information to support regional impacts assessment and adaptation options appraisal. However, questions continue to surround how best to apply these scenarios in a practical context, and whether the added complexity and computational burden leads to more robust decision-making. This paper provides an overview of recent efforts in the UK to 'bench-test' frameworks for employing probabilistic projections ahead of the release of the next generation, UKCIP08 projections (in November 2008). This is involving close collaboration between government agencies, research and stakeholder communities. Three examples will be cited to illustrate how probabilistic projections are already informing decisions about future flood risk management in London, water resource planning in trial river basins, and assessments of risks from rising water temperatures to Atlantic salmon stocks in southern England. When compared with conventional deterministic scenarios, ensemble projections allow exploration of a wider range of management options and highlight timescales for implementing adaptation measures. Users of probabilistic scenarios must keep in mind that other uncertainties (e.g., due to impacts model structure and parameterisation) should be handled in an equally rigorous way to those arising from climate models and emission scenarios. Finally, it is noted that a commitment to long-term monitoring is also critical for tracking environmental change, testing model projections, and for evaluating the success (or not) of any scenario-led interventions.

Mode identification using stochastic hybrid models with applications to conflict detection and resolution

NASA Astrophysics Data System (ADS)

Naseri Kouzehgarani, Asal

2009-12-01

Most models of aircraft trajectories are non-linear and stochastic in nature; and their internal parameters are often poorly defined. The ability to model, simulate and analyze realistic air traffic management conflict detection scenarios in a scalable, composable, multi-aircraft fashion is an extremely difficult endeavor. Accurate techniques for aircraft mode detection are critical in order to enable the precise projection of aircraft conflicts, and for the enactment of altitude separation resolution strategies. Conflict detection is an inherently probabilistic endeavor; our ability to detect conflicts in a timely and accurate manner over a fixed time horizon is traded off against the increased human workload created by false alarms---that is, situations that would not develop into an actual conflict, or would resolve naturally in the appropriate time horizon-thereby introducing a measure of probabilistic uncertainty in any decision aid fashioned to assist air traffic controllers. The interaction of the continuous dynamics of the aircraft, used for prediction purposes, with the discrete conflict detection logic gives rise to the hybrid nature of the overall system. The introduction of the probabilistic element, common to decision alerting and aiding devices, places the conflict detection and resolution problem in the domain of probabilistic hybrid phenomena. A hidden Markov model (HMM) has two stochastic components: a finite-state Markov chain and a finite set of output probability distributions. In other words an unobservable stochastic process (hidden) that can only be observed through another set of stochastic processes that generate the sequence of observations. The problem of self separation in distributed air traffic management reduces to the ability of aircraft to communicate state information to neighboring aircraft, as well as model the evolution of aircraft trajectories between communications, in the presence of probabilistic uncertain dynamics as well as partially observable and uncertain data. We introduce the Hybrid Hidden Markov Modeling (HHMM) formalism to enable the prediction of the stochastic aircraft states (and thus, potential conflicts), by combining elements of the probabilistic timed input output automaton and the partially observable Markov decision process frameworks, along with the novel addition of a Markovian scheduler to remove the non-deterministic elements arising from the enabling of several actions simultaneously. Comparisons of aircraft in level, climbing/descending and turning flight are performed, and unknown flight track data is evaluated probabilistically against the tuned model in order to assess the effectiveness of the model in detecting the switch between multiple flight modes for a given aircraft. This also allows for the generation of probabilistic distribution over the execution traces of the hybrid hidden Markov model, which then enables the prediction of the states of aircraft based on partially observable and uncertain data. Based on the composition properties of the HHMM, we study a decentralized air traffic system where aircraft are moving along streams and can perform cruise, accelerate, climb and turn maneuvers. We develop a common decentralized policy for conflict avoidance with spatially distributed agents (aircraft in the sky) and assure its safety properties via correctness proofs.

System Level Uncertainty Assessment for Collaborative RLV Design

NASA Technical Reports Server (NTRS)

Charania, A. C.; Bradford, John E.; Olds, John R.; Graham, Matthew

2002-01-01

A collaborative design process utilizing Probabilistic Data Assessment (PDA) is showcased. Given the limitation of financial resources by both the government and industry, strategic decision makers need more than just traditional point designs, they need to be aware of the likelihood of these future designs to meet their objectives. This uncertainty, an ever-present character in the design process, can be embraced through a probabilistic design environment. A conceptual design process is presented that encapsulates the major engineering disciplines for a Third Generation Reusable Launch Vehicle (RLV). Toolsets consist of aerospace industry standard tools in disciplines such as trajectory, propulsion, mass properties, cost, operations, safety, and economics. Variations of the design process are presented that use different fidelities of tools. The disciplinary engineering models are used in a collaborative engineering framework utilizing Phoenix Integration's ModelCenter and AnalysisServer environment. These tools allow the designer to join disparate models and simulations together in a unified environment wherein each discipline can interact with any other discipline. The design process also uses probabilistic methods to generate the system level output metrics of interest for a RLV conceptual design. The specific system being examined is the Advanced Concept Rocket Engine 92 (ACRE-92) RLV. Previous experience and knowledge (in terms of input uncertainty distributions from experts and modeling and simulation codes) can be coupled with Monte Carlo processes to best predict the chances of program success.

Near Real-Time Probabilistic Damage Diagnosis Using Surrogate Modeling and High Performance Computing

NASA Technical Reports Server (NTRS)

Warner, James E.; Zubair, Mohammad; Ranjan, Desh

2017-01-01

This work investigates novel approaches to probabilistic damage diagnosis that utilize surrogate modeling and high performance computing (HPC) to achieve substantial computational speedup. Motivated by Digital Twin, a structural health management (SHM) paradigm that integrates vehicle-specific characteristics with continual in-situ damage diagnosis and prognosis, the methods studied herein yield near real-time damage assessments that could enable monitoring of a vehicle's health while it is operating (i.e. online SHM). High-fidelity modeling and uncertainty quantification (UQ), both critical to Digital Twin, are incorporated using finite element method simulations and Bayesian inference, respectively. The crux of the proposed Bayesian diagnosis methods, however, is the reformulation of the numerical sampling algorithms (e.g. Markov chain Monte Carlo) used to generate the resulting probabilistic damage estimates. To this end, three distinct methods are demonstrated for rapid sampling that utilize surrogate modeling and exploit various degrees of parallelism for leveraging HPC. The accuracy and computational efficiency of the methods are compared on the problem of strain-based crack identification in thin plates. While each approach has inherent problem-specific strengths and weaknesses, all approaches are shown to provide accurate probabilistic damage diagnoses and several orders of magnitude computational speedup relative to a baseline Bayesian diagnosis implementation.

A comprehensive dwelling unit choice model accommodating psychological constructs within a search strategy for consideration set formation.

DOT National Transportation Integrated Search

2015-12-01

This study adopts a dwelling unit level of analysis and considers a probabilistic choice set generation approach for residential choice modeling. In doing so, we accommodate the fact that housing choices involve both characteristics of the dwelling u...

bayesPop: Probabilistic Population Projections

PubMed Central

Ševčíková, Hana; Raftery, Adrian E.

2016-01-01

We describe bayesPop, an R package for producing probabilistic population projections for all countries. This uses probabilistic projections of total fertility and life expectancy generated by Bayesian hierarchical models. It produces a sample from the joint posterior predictive distribution of future age- and sex-specific population counts, fertility rates and mortality rates, as well as future numbers of births and deaths. It provides graphical ways of summarizing this information, including trajectory plots and various kinds of probabilistic population pyramids. An expression language is introduced which allows the user to produce the predictive distribution of a wide variety of derived population quantities, such as the median age or the old age dependency ratio. The package produces aggregated projections for sets of countries, such as UN regions or trading blocs. The methodology has been used by the United Nations to produce their most recent official population projections for all countries, published in the World Population Prospects. PMID:28077933

Unifying Model-Based and Reactive Programming within a Model-Based Executive

NASA Technical Reports Server (NTRS)

Williams, Brian C.; Gupta, Vineet; Norvig, Peter (Technical Monitor)

1999-01-01

Real-time, model-based, deduction has recently emerged as a vital component in AI's tool box for developing highly autonomous reactive systems. Yet one of the current hurdles towards developing model-based reactive systems is the number of methods simultaneously employed, and their corresponding melange of programming and modeling languages. This paper offers an important step towards unification. We introduce RMPL, a rich modeling language that combines probabilistic, constraint-based modeling with reactive programming constructs, while offering a simple semantics in terms of hidden state Markov processes. We introduce probabilistic, hierarchical constraint automata (PHCA), which allow Markov processes to be expressed in a compact representation that preserves the modularity of RMPL programs. Finally, a model-based executive, called Reactive Burton is described that exploits this compact encoding to perform efficIent simulation, belief state update and control sequence generation.

The exposure of Sydney (Australia) to earthquake-generated tsunamis, storms and sea level rise: a probabilistic multi-hazard approach

PubMed Central

Dall'Osso, F.; Dominey-Howes, D.; Moore, C.; Summerhayes, S.; Withycombe, G.

2014-01-01

Approximately 85% of Australia's population live along the coastal fringe, an area with high exposure to extreme inundations such as tsunamis. However, to date, no Probabilistic Tsunami Hazard Assessments (PTHA) that include inundation have been published for Australia. This limits the development of appropriate risk reduction measures by decision and policy makers. We describe our PTHA undertaken for the Sydney metropolitan area. Using the NOAA NCTR model MOST (Method for Splitting Tsunamis), we simulate 36 earthquake-generated tsunamis with annual probabilities of 1:100, 1:1,000 and 1:10,000, occurring under present and future predicted sea level conditions. For each tsunami scenario we generate a high-resolution inundation map of the maximum water level and flow velocity, and we calculate the exposure of buildings and critical infrastructure. Results indicate that exposure to earthquake-generated tsunamis is relatively low for present events, but increases significantly with higher sea level conditions. The probabilistic approach allowed us to undertake a comparison with an existing storm surge hazard assessment. Interestingly, the exposure to all the simulated tsunamis is significantly lower than that for the 1:100 storm surge scenarios, under the same initial sea level conditions. The results have significant implications for multi-risk and emergency management in Sydney. PMID:25492514

The exposure of Sydney (Australia) to earthquake-generated tsunamis, storms and sea level rise: a probabilistic multi-hazard approach.

PubMed

Dall'Osso, F; Dominey-Howes, D; Moore, C; Summerhayes, S; Withycombe, G

2014-12-10

Approximately 85% of Australia's population live along the coastal fringe, an area with high exposure to extreme inundations such as tsunamis. However, to date, no Probabilistic Tsunami Hazard Assessments (PTHA) that include inundation have been published for Australia. This limits the development of appropriate risk reduction measures by decision and policy makers. We describe our PTHA undertaken for the Sydney metropolitan area. Using the NOAA NCTR model MOST (Method for Splitting Tsunamis), we simulate 36 earthquake-generated tsunamis with annual probabilities of 1:100, 1:1,000 and 1:10,000, occurring under present and future predicted sea level conditions. For each tsunami scenario we generate a high-resolution inundation map of the maximum water level and flow velocity, and we calculate the exposure of buildings and critical infrastructure. Results indicate that exposure to earthquake-generated tsunamis is relatively low for present events, but increases significantly with higher sea level conditions. The probabilistic approach allowed us to undertake a comparison with an existing storm surge hazard assessment. Interestingly, the exposure to all the simulated tsunamis is significantly lower than that for the 1:100 storm surge scenarios, under the same initial sea level conditions. The results have significant implications for multi-risk and emergency management in Sydney.

Mixing zone and drinking water intake dilution factor and wastewater generation distributions to enable probabilistic assessment of down-the-drain consumer product chemicals in the U.S.

PubMed

Kapo, Katherine E; McDonough, Kathleen; Federle, Thomas; Dyer, Scott; Vamshi, Raghu

2015-06-15

Environmental exposure and associated ecological risk related to down-the-drain chemicals discharged by municipal wastewater treatment plants (WWTPs) are strongly influenced by in-stream dilution of receiving waters which varies by geography, flow conditions and upstream wastewater inputs. The iSTREEM® model (American Cleaning Institute, Washington D.C.) was utilized to determine probabilistic distributions for no decay and decay-based dilution factors in mean annual and low (7Q10) flow conditions. The dilution factors derived in this study are "combined" dilution factors which account for both hydrologic dilution and cumulative upstream effluent contributions that will differ depending on the rate of in-stream decay due to biodegradation, volatilization, sorption, etc. for the chemical being evaluated. The median dilution factors estimated in this study (based on various in-stream decay rates from zero decay to a 1h half-life) for WWTP mixing zones dominated by domestic wastewater flow ranged from 132 to 609 at mean flow and 5 to 25 at low flow, while median dilution factors at drinking water intakes (mean flow) ranged from 146 to 2×10(7) depending on the in-stream decay rate. WWTPs within the iSTREEM® model were used to generate a distribution of per capita wastewater generated in the U.S. The dilution factor and per capita wastewater generation distributions developed by this work can be used to conduct probabilistic exposure assessments for down-the-drain chemicals in influent wastewater, wastewater treatment plant mixing zones and at drinking water intakes in the conterminous U.S. In addition, evaluation of types and abundance of U.S. wastewater treatment processes provided insight into treatment trends and the flow volume treated by each type of process. Moreover, removal efficiencies of chemicals can differ by treatment type. Hence, the availability of distributions for per capita wastewater production, treatment type, and dilution factors at a national level provides a series of practical and powerful tools for building probabilistic exposure models. Copyright © 2015 Elsevier B.V. All rights reserved.

From Cyclone Tracks to the Costs of European Winter Storms: A Probabilistic Loss Assessment Model

NASA Astrophysics Data System (ADS)

Orwig, K.; Renggli, D.; Corti, T.; Reese, S.; Wueest, M.; Viktor, E.; Zimmerli, P.

2014-12-01

European winter storms cause billions of dollars of insured losses every year. Therefore, it is essential to understand potential impacts of future events, and the role reinsurance can play to mitigate the losses. The authors will present an overview on natural catastrophe risk assessment modeling in the reinsurance industry, and the development of a new innovative approach for modeling the risk associated with European winter storms.The new innovative approach includes the development of physically meaningful probabilistic (i.e. simulated) events for European winter storm loss assessment. The meteorological hazard component of the new model is based on cyclone and windstorm tracks identified in the 20thCentury Reanalysis data. The knowledge of the evolution of winter storms both in time and space allows the physically meaningful perturbation of historical event properties (e.g. track, intensity, etc.). The perturbation includes a random element but also takes the local climatology and the evolution of the historical event into account.The low-resolution wind footprints taken from the 20thCentury Reanalysis are processed by a statistical-dynamical downscaling to generate high-resolution footprints for both the simulated and historical events. Downscaling transfer functions are generated using ENSEMBLES regional climate model data. The result is a set of reliable probabilistic events representing thousands of years. The event set is then combined with country and site-specific vulnerability functions and detailed market- or client-specific information to compute annual expected losses.

Encoding probabilistic brain atlases using Bayesian inference.

PubMed

Van Leemput, Koen

2009-06-01

This paper addresses the problem of creating probabilistic brain atlases from manually labeled training data. Probabilistic atlases are typically constructed by counting the relative frequency of occurrence of labels in corresponding locations across the training images. However, such an "averaging" approach generalizes poorly to unseen cases when the number of training images is limited, and provides no principled way of aligning the training datasets using deformable registration. In this paper, we generalize the generative image model implicitly underlying standard "average" atlases, using mesh-based representations endowed with an explicit deformation model. Bayesian inference is used to infer the optimal model parameters from the training data, leading to a simultaneous group-wise registration and atlas estimation scheme that encompasses standard averaging as a special case. We also use Bayesian inference to compare alternative atlas models in light of the training data, and show how this leads to a data compression problem that is intuitive to interpret and computationally feasible. Using this technique, we automatically determine the optimal amount of spatial blurring, the best deformation field flexibility, and the most compact mesh representation. We demonstrate, using 2-D training datasets, that the resulting models are better at capturing the structure in the training data than conventional probabilistic atlases. We also present experiments of the proposed atlas construction technique in 3-D, and show the resulting atlases' potential in fully-automated, pulse sequence-adaptive segmentation of 36 neuroanatomical structures in brain MRI scans.

Theory Learning as Stochastic Search in the Language of Thought

ERIC Educational Resources Information Center

Ullman, Tomer D.; Goodman, Noah D.; Tenenbaum, Joshua B.

2012-01-01

We present an algorithmic model for the development of children's intuitive theories within a hierarchical Bayesian framework, where theories are described as sets of logical laws generated by a probabilistic context-free grammar. We contrast our approach with connectionist and other emergentist approaches to modeling cognitive development. While…

Obtaining Accurate Probabilities Using Classifier Calibration

ERIC Educational Resources Information Center

Pakdaman Naeini, Mahdi

2016-01-01

Learning probabilistic classification and prediction models that generate accurate probabilities is essential in many prediction and decision-making tasks in machine learning and data mining. One way to achieve this goal is to post-process the output of classification models to obtain more accurate probabilities. These post-processing methods are…

Enhancing Flood Prediction Reliability Using Bayesian Model Averaging

NASA Astrophysics Data System (ADS)

Liu, Z.; Merwade, V.

2017-12-01

Uncertainty analysis is an indispensable part of modeling the hydrology and hydrodynamics of non-idealized environmental systems. Compared to reliance on prediction from one model simulation, using on ensemble of predictions that consider uncertainty from different sources is more reliable. In this study, Bayesian model averaging (BMA) is applied to Black River watershed in Arkansas and Missouri by combining multi-model simulations to get reliable deterministic water stage and probabilistic inundation extent predictions. The simulation ensemble is generated from 81 LISFLOOD-FP subgrid model configurations that include uncertainty from channel shape, channel width, channel roughness and discharge. Model simulation outputs are trained with observed water stage data during one flood event, and BMA prediction ability is validated for another flood event. Results from this study indicate that BMA does not always outperform all members in the ensemble, but it provides relatively robust deterministic flood stage predictions across the basin. Station based BMA (BMA_S) water stage prediction has better performance than global based BMA (BMA_G) prediction which is superior to the ensemble mean prediction. Additionally, high-frequency flood inundation extent (probability greater than 60%) in BMA_G probabilistic map is more accurate than the probabilistic flood inundation extent based on equal weights.

Generative diffeomorphic modelling of large MRI data sets for probabilistic template construction.

PubMed

Blaiotta, Claudia; Freund, Patrick; Cardoso, M Jorge; Ashburner, John

2018-02-01

In this paper we present a hierarchical generative model of medical image data, which can capture simultaneously the variability of both signal intensity and anatomical shapes across large populations. Such a model has a direct application for learning average-shaped probabilistic tissue templates in a fully automated manner. While in principle the generality of the proposed Bayesian approach makes it suitable to address a wide range of medical image computing problems, our work focuses primarily on neuroimaging applications. In particular we validate the proposed method on both real and synthetic brain MR scans including the cervical cord and demonstrate that it yields accurate alignment of brain and spinal cord structures, as compared to state-of-the-art tools for medical image registration. At the same time we illustrate how the resulting tissue probability maps can readily be used to segment, bias correct and spatially normalise unseen data, which are all crucial pre-processing steps for MR imaging studies. Copyright © 2017 The Authors. Published by Elsevier Inc. All rights reserved.

Testing for ontological errors in probabilistic forecasting models of natural systems

PubMed Central

Marzocchi, Warner; Jordan, Thomas H.

2014-01-01

Probabilistic forecasting models describe the aleatory variability of natural systems as well as our epistemic uncertainty about how the systems work. Testing a model against observations exposes ontological errors in the representation of a system and its uncertainties. We clarify several conceptual issues regarding the testing of probabilistic forecasting models for ontological errors: the ambiguity of the aleatory/epistemic dichotomy, the quantification of uncertainties as degrees of belief, the interplay between Bayesian and frequentist methods, and the scientific pathway for capturing predictability. We show that testability of the ontological null hypothesis derives from an experimental concept, external to the model, that identifies collections of data, observed and not yet observed, that are judged to be exchangeable when conditioned on a set of explanatory variables. These conditional exchangeability judgments specify observations with well-defined frequencies. Any model predicting these behaviors can thus be tested for ontological error by frequentist methods; e.g., using P values. In the forecasting problem, prior predictive model checking, rather than posterior predictive checking, is desirable because it provides more severe tests. We illustrate experimental concepts using examples from probabilistic seismic hazard analysis. Severe testing of a model under an appropriate set of experimental concepts is the key to model validation, in which we seek to know whether a model replicates the data-generating process well enough to be sufficiently reliable for some useful purpose, such as long-term seismic forecasting. Pessimistic views of system predictability fail to recognize the power of this methodology in separating predictable behaviors from those that are not. PMID:25097265

Generating probabilistic Boolean networks from a prescribed transition probability matrix.

PubMed

Ching, W-K; Chen, X; Tsing, N-K

2009-11-01

Probabilistic Boolean networks (PBNs) have received much attention in modeling genetic regulatory networks. A PBN can be regarded as a Markov chain process and is characterised by a transition probability matrix. In this study, the authors propose efficient algorithms for constructing a PBN when its transition probability matrix is given. The complexities of the algorithms are also analysed. This is an interesting inverse problem in network inference using steady-state data. The problem is important as most microarray data sets are assumed to be obtained from sampling the steady-state.

Systems and methods that generate height map models for efficient three dimensional reconstruction from depth information

DOEpatents

Frahm, Jan-Michael; Pollefeys, Marc Andre Leon; Gallup, David Robert

2015-12-08

Methods of generating a three dimensional representation of an object in a reference plane from a depth map including distances from a reference point to pixels in an image of the object taken from a reference point. Weights are assigned to respective voxels in a three dimensional grid along rays extending from the reference point through the pixels in the image based on the distances in the depth map from the reference point to the respective pixels, and a height map including an array of height values in the reference plane is formed based on the assigned weights. An n-layer height map may be constructed by generating a probabilistic occupancy grid for the voxels and forming an n-dimensional height map comprising an array of layer height values in the reference plane based on the probabilistic occupancy grid.

Managing flowback and produced water from hydraulic fracturing under stochastic environment

NASA Astrophysics Data System (ADS)

Zhang, X.; Sun, A. Y.; Duncan, I. J.; Vesselinov, V. V.

2017-12-01

A large volume of wastewater is being generated from hydraulic fracturing in shale gas plays, including flowback and produced water. The produced wastewater in terms of its quantity and quality has become one of the main environmental problems facing shale gas industries worldwide. Cost-effective planning and management of flowback and produced water is highly desirable. Careful choice of treatment, disposal, and reuse options can lower costs and reduce potential environmental impacts. To handle the recourse issue in decision-making, a two-stage stochastic management model is developed to provide optimal alternatives for fracturing wastewater management. The proposed model is capable of prompting corrective actions to allow decision makers to adjust the pre-defined management strategies. By using this two-stage model, potential penalties arising from decision infeasibility can be minimized. The applicability of the proposed model is demonstrated using a representative synthetic example, in which tradeoffs between economic and environmental goals are quantified. This approach can generate informed defensible decisions for shale gas wastewater management. In addition, probabilistic and non-probabilistic uncertainties are effectively addressed.

Multi-model ensemble hydrologic prediction using Bayesian model averaging

NASA Astrophysics Data System (ADS)

Duan, Qingyun; Ajami, Newsha K.; Gao, Xiaogang; Sorooshian, Soroosh

2007-05-01

Multi-model ensemble strategy is a means to exploit the diversity of skillful predictions from different models. This paper studies the use of Bayesian model averaging (BMA) scheme to develop more skillful and reliable probabilistic hydrologic predictions from multiple competing predictions made by several hydrologic models. BMA is a statistical procedure that infers consensus predictions by weighing individual predictions based on their probabilistic likelihood measures, with the better performing predictions receiving higher weights than the worse performing ones. Furthermore, BMA provides a more reliable description of the total predictive uncertainty than the original ensemble, leading to a sharper and better calibrated probability density function (PDF) for the probabilistic predictions. In this study, a nine-member ensemble of hydrologic predictions was used to test and evaluate the BMA scheme. This ensemble was generated by calibrating three different hydrologic models using three distinct objective functions. These objective functions were chosen in a way that forces the models to capture certain aspects of the hydrograph well (e.g., peaks, mid-flows and low flows). Two sets of numerical experiments were carried out on three test basins in the US to explore the best way of using the BMA scheme. In the first set, a single set of BMA weights was computed to obtain BMA predictions, while the second set employed multiple sets of weights, with distinct sets corresponding to different flow intervals. In both sets, the streamflow values were transformed using Box-Cox transformation to ensure that the probability distribution of the prediction errors is approximately Gaussian. A split sample approach was used to obtain and validate the BMA predictions. The test results showed that BMA scheme has the advantage of generating more skillful and equally reliable probabilistic predictions than original ensemble. The performance of the expected BMA predictions in terms of daily root mean square error (DRMS) and daily absolute mean error (DABS) is generally superior to that of the best individual predictions. Furthermore, the BMA predictions employing multiple sets of weights are generally better than those using single set of weights.

Modeling the Evolution of Beliefs Using an Attentional Focus Mechanism

PubMed Central

Marković, Dimitrije; Gläscher, Jan; Bossaerts, Peter; O’Doherty, John; Kiebel, Stefan J.

2015-01-01

For making decisions in everyday life we often have first to infer the set of environmental features that are relevant for the current task. Here we investigated the computational mechanisms underlying the evolution of beliefs about the relevance of environmental features in a dynamical and noisy environment. For this purpose we designed a probabilistic Wisconsin card sorting task (WCST) with belief solicitation, in which subjects were presented with stimuli composed of multiple visual features. At each moment in time a particular feature was relevant for obtaining reward, and participants had to infer which feature was relevant and report their beliefs accordingly. To test the hypothesis that attentional focus modulates the belief update process, we derived and fitted several probabilistic and non-probabilistic behavioral models, which either incorporate a dynamical model of attentional focus, in the form of a hierarchical winner-take-all neuronal network, or a diffusive model, without attention-like features. We used Bayesian model selection to identify the most likely generative model of subjects’ behavior and found that attention-like features in the behavioral model are essential for explaining subjects’ responses. Furthermore, we demonstrate a method for integrating both connectionist and Bayesian models of decision making within a single framework that allowed us to infer hidden belief processes of human subjects. PMID:26495984

Effects of sample survey design on the accuracy of classification tree models in species distribution models

USGS Publications Warehouse

Edwards, T.C.; Cutler, D.R.; Zimmermann, N.E.; Geiser, L.; Moisen, Gretchen G.

2006-01-01

We evaluated the effects of probabilistic (hereafter DESIGN) and non-probabilistic (PURPOSIVE) sample surveys on resultant classification tree models for predicting the presence of four lichen species in the Pacific Northwest, USA. Models derived from both survey forms were assessed using an independent data set (EVALUATION). Measures of accuracy as gauged by resubstitution rates were similar for each lichen species irrespective of the underlying sample survey form. Cross-validation estimates of prediction accuracies were lower than resubstitution accuracies for all species and both design types, and in all cases were closer to the true prediction accuracies based on the EVALUATION data set. We argue that greater emphasis should be placed on calculating and reporting cross-validation accuracy rates rather than simple resubstitution accuracy rates. Evaluation of the DESIGN and PURPOSIVE tree models on the EVALUATION data set shows significantly lower prediction accuracy for the PURPOSIVE tree models relative to the DESIGN models, indicating that non-probabilistic sample surveys may generate models with limited predictive capability. These differences were consistent across all four lichen species, with 11 of the 12 possible species and sample survey type comparisons having significantly lower accuracy rates. Some differences in accuracy were as large as 50%. The classification tree structures also differed considerably both among and within the modelled species, depending on the sample survey form. Overlap in the predictor variables selected by the DESIGN and PURPOSIVE tree models ranged from only 20% to 38%, indicating the classification trees fit the two evaluated survey forms on different sets of predictor variables. The magnitude of these differences in predictor variables throws doubt on ecological interpretation derived from prediction models based on non-probabilistic sample surveys. ?? 2006 Elsevier B.V. All rights reserved.

Stochastic Simulation and Forecast of Hydrologic Time Series Based on Probabilistic Chaos Expansion

NASA Astrophysics Data System (ADS)

Li, Z.; Ghaith, M.

2017-12-01

Hydrological processes are characterized by many complex features, such as nonlinearity, dynamics and uncertainty. How to quantify and address such complexities and uncertainties has been a challenging task for water engineers and managers for decades. To support robust uncertainty analysis, an innovative approach for the stochastic simulation and forecast of hydrologic time series is developed is this study. Probabilistic Chaos Expansions (PCEs) are established through probabilistic collocation to tackle uncertainties associated with the parameters of traditional hydrological models. The uncertainties are quantified in model outputs as Hermite polynomials with regard to standard normal random variables. Sequentially, multivariate analysis techniques are used to analyze the complex nonlinear relationships between meteorological inputs (e.g., temperature, precipitation, evapotranspiration, etc.) and the coefficients of the Hermite polynomials. With the established relationships between model inputs and PCE coefficients, forecasts of hydrologic time series can be generated and the uncertainties in the future time series can be further tackled. The proposed approach is demonstrated using a case study in China and is compared to a traditional stochastic simulation technique, the Markov-Chain Monte-Carlo (MCMC) method. Results show that the proposed approach can serve as a reliable proxy to complicated hydrological models. It can provide probabilistic forecasting in a more computationally efficient manner, compared to the traditional MCMC method. This work provides technical support for addressing uncertainties associated with hydrological modeling and for enhancing the reliability of hydrological modeling results. Applications of the developed approach can be extended to many other complicated geophysical and environmental modeling systems to support the associated uncertainty quantification and risk analysis.

Addressing the Hard Factors for Command File Errors by Probabilistic Reasoning

NASA Technical Reports Server (NTRS)

Meshkat, Leila; Bryant, Larry

2014-01-01

Command File Errors (CFE) are managed using standard risk management approaches at the Jet Propulsion Laboratory. Over the last few years, more emphasis has been made on the collection, organization, and analysis of these errors for the purpose of reducing the CFE rates. More recently, probabilistic modeling techniques have been used for more in depth analysis of the perceived error rates of the DAWN mission and for managing the soft factors in the upcoming phases of the mission. We broadly classify the factors that can lead to CFE's as soft factors, which relate to the cognition of the operators and hard factors which relate to the Mission System which is composed of the hardware, software and procedures used for the generation, verification & validation and execution of commands. The focus of this paper is to use probabilistic models that represent multiple missions at JPL to determine the root cause and sensitivities of the various components of the mission system and develop recommendations and techniques for addressing them. The customization of these multi-mission models to a sample interplanetary spacecraft is done for this purpose.

Probabilistic Modeling of High-Temperature Material Properties of a 5-Harness 0/90 Sylramic Fiber/ CVI-SiC/ MI-SiC Woven Composite

NASA Technical Reports Server (NTRS)

Nagpal, Vinod K.; Tong, Michael; Murthy, P. L. N.; Mital, Subodh

1998-01-01

An integrated probabilistic approach has been developed to assess composites for high temperature applications. This approach was used to determine thermal and mechanical properties and their probabilistic distributions of a 5-harness 0/90 Sylramic fiber/CVI-SiC/Mi-SiC woven Ceramic Matrix Composite (CMC) at high temperatures. The purpose of developing this approach was to generate quantitative probabilistic information on this CMC to help complete the evaluation for its potential application for HSCT combustor liner. This approach quantified the influences of uncertainties inherent in constituent properties called primitive variables on selected key response variables of the CMC at 2200 F. The quantitative information is presented in the form of Cumulative Density Functions (CDFs). Probability Density Functions (PDFS) and primitive variable sensitivities on response. Results indicate that the scatters in response variables were reduced by 30-50% when the uncertainties in the primitive variables, which showed the most influence, were reduced by 50%.

Multi-agent simulation of generation expansion in electricity markets.

DOE Office of Scientific and Technical Information (OSTI.GOV)

Botterud, A; Mahalik, M. R.; Veselka, T. D.

2007-06-01

We present a new multi-agent model of generation expansion in electricity markets. The model simulates generation investment decisions of decentralized generating companies (GenCos) interacting in a complex, multidimensional environment. A probabilistic dispatch algorithm calculates prices and profits for new candidate units in different future states of the system. Uncertainties in future load, hydropower conditions, and competitors actions are represented in a scenario tree, and decision analysis is used to identify the optimal expansion decision for each individual GenCo. We test the model using real data for the Korea power system under different assumptions about market design, market concentration, and GenCo'smore » assumed expectations about their competitors investment decisions.« less

Using probabilistic theory to develop interpretation guidelines for Y-STR profiles.

PubMed

Taylor, Duncan; Bright, Jo-Anne; Buckleton, John

2016-03-01

Y-STR profiling makes up a small but important proportion of forensic DNA casework. Often Y-STR profiles are used when autosomal profiling has failed to yield an informative result. Consequently Y-STR profiles are often from the most challenging samples. In addition to these points, Y-STR loci are linked, meaning that evaluation of haplotype probabilities are either based on overly simplified counting methods or computationally costly genetic models, neither of which extend well to the evaluation of mixed Y-STR data. For all of these reasons Y-STR data analysis has not seen the same advances as autosomal STR data. We present here a probabilistic model for the interpretation of Y-STR data. Due to the fact that probabilistic systems for Y-STR data are still some way from reaching active casework, we also describe how data can be analysed in a continuous way to generate interpretational thresholds and guidelines. Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.

A probabilistic cellular automata model for the dynamics of a population driven by logistic growth and weak Allee effect

NASA Astrophysics Data System (ADS)

Mendonça, J. R. G.

2018-04-01

We propose and investigate a one-parameter probabilistic mixture of one-dimensional elementary cellular automata under the guise of a model for the dynamics of a single-species unstructured population with nonoverlapping generations in which individuals have smaller probability of reproducing and surviving in a crowded neighbourhood but also suffer from isolation and dispersal. Remarkably, the first-order mean field approximation to the dynamics of the model yields a cubic map containing terms representing both logistic and weak Allee effects. The model has a single absorbing state devoid of individuals, but depending on the reproduction and survival probabilities can achieve a stable population. We determine the critical probability separating these two phases and find that the phase transition between them is in the directed percolation universality class of critical behaviour.

A probabilistic approach to photovoltaic generator performance prediction

NASA Astrophysics Data System (ADS)

Khallat, M. A.; Rahman, S.

1986-09-01

A method for predicting the performance of a photovoltaic (PV) generator based on long term climatological data and expected cell performance is described. The equations for cell model formulation are provided. Use of the statistical model for characterizing the insolation level is discussed. The insolation data is fitted to appropriate probability distribution functions (Weibull, beta, normal). The probability distribution functions are utilized to evaluate the capacity factors of PV panels or arrays. An example is presented revealing the applicability of the procedure.

Learning a Generative Probabilistic Grammar of Experience: A Process-Level Model of Language Acquisition

ERIC Educational Resources Information Center

Kolodny, Oren; Lotem, Arnon; Edelman, Shimon

2015-01-01

We introduce a set of biologically and computationally motivated design choices for modeling the learning of language, or of other types of sequential, hierarchically structured experience and behavior, and describe an implemented system that conforms to these choices and is capable of unsupervised learning from raw natural-language corpora. Given…

A look-ahead probabilistic contingency analysis framework incorporating smart sampling techniques

DOE Office of Scientific and Technical Information (OSTI.GOV)

Chen, Yousu; Etingov, Pavel V.; Ren, Huiying

2016-07-18

This paper describes a framework of incorporating smart sampling techniques in a probabilistic look-ahead contingency analysis application. The predictive probabilistic contingency analysis helps to reflect the impact of uncertainties caused by variable generation and load on potential violations of transmission limits.

A Probabilistic Feature Map-Based Localization System Using a Monocular Camera.

PubMed

Kim, Hyungjin; Lee, Donghwa; Oh, Taekjun; Choi, Hyun-Taek; Myung, Hyun

2015-08-31

Image-based localization is one of the most widely researched localization techniques in the robotics and computer vision communities. As enormous image data sets are provided through the Internet, many studies on estimating a location with a pre-built image-based 3D map have been conducted. Most research groups use numerous image data sets that contain sufficient features. In contrast, this paper focuses on image-based localization in the case of insufficient images and features. A more accurate localization method is proposed based on a probabilistic map using 3D-to-2D matching correspondences between a map and a query image. The probabilistic feature map is generated in advance by probabilistic modeling of the sensor system as well as the uncertainties of camera poses. Using the conventional PnP algorithm, an initial camera pose is estimated on the probabilistic feature map. The proposed algorithm is optimized from the initial pose by minimizing Mahalanobis distance errors between features from the query image and the map to improve accuracy. To verify that the localization accuracy is improved, the proposed algorithm is compared with the conventional algorithm in a simulation and realenvironments.

A Probabilistic Feature Map-Based Localization System Using a Monocular Camera

PubMed Central

Kim, Hyungjin; Lee, Donghwa; Oh, Taekjun; Choi, Hyun-Taek; Myung, Hyun

2015-01-01

Image-based localization is one of the most widely researched localization techniques in the robotics and computer vision communities. As enormous image data sets are provided through the Internet, many studies on estimating a location with a pre-built image-based 3D map have been conducted. Most research groups use numerous image data sets that contain sufficient features. In contrast, this paper focuses on image-based localization in the case of insufficient images and features. A more accurate localization method is proposed based on a probabilistic map using 3D-to-2D matching correspondences between a map and a query image. The probabilistic feature map is generated in advance by probabilistic modeling of the sensor system as well as the uncertainties of camera poses. Using the conventional PnP algorithm, an initial camera pose is estimated on the probabilistic feature map. The proposed algorithm is optimized from the initial pose by minimizing Mahalanobis distance errors between features from the query image and the map to improve accuracy. To verify that the localization accuracy is improved, the proposed algorithm is compared with the conventional algorithm in a simulation and realenvironments. PMID:26404284

Disruption of the Right Temporoparietal Junction Impairs Probabilistic Belief Updating.

PubMed

Mengotti, Paola; Dombert, Pascasie L; Fink, Gereon R; Vossel, Simone

2017-05-31

Generating and updating probabilistic models of the environment is a fundamental modus operandi of the human brain. Although crucial for various cognitive functions, the neural mechanisms of these inference processes remain to be elucidated. Here, we show the causal involvement of the right temporoparietal junction (rTPJ) in updating probabilistic beliefs and we provide new insights into the chronometry of the process by combining online transcranial magnetic stimulation (TMS) with computational modeling of behavioral responses. Female and male participants performed a modified location-cueing paradigm, where false information about the percentage of cue validity (%CV) was provided in half of the experimental blocks to prompt updating of prior expectations. Online double-pulse TMS over rTPJ 300 ms (but not 50 ms) after target appearance selectively decreased participants' updating of false prior beliefs concerning %CV, reflected in a decreased learning rate of a Rescorla-Wagner model. Online TMS over rTPJ also impacted on participants' explicit beliefs, causing them to overestimate %CV. These results confirm the involvement of rTPJ in updating of probabilistic beliefs, thereby advancing our understanding of this area's function during cognitive processing. SIGNIFICANCE STATEMENT Contemporary views propose that the brain maintains probabilistic models of the world to minimize surprise about sensory inputs. Here, we provide evidence that the right temporoparietal junction (rTPJ) is causally involved in this process. Because neuroimaging has suggested that rTPJ is implicated in divergent cognitive domains, the demonstration of an involvement in updating internal models provides a novel unifying explanation for these findings. We used computational modeling to characterize how participants change their beliefs after new observations. By interfering with rTPJ activity through online transcranial magnetic stimulation, we showed that participants were less able to update prior beliefs with TMS delivered at 300 ms after target onset. Copyright © 2017 the authors 0270-6474/17/375419-10$15.00/0.

The Smoothed Dirichlet Distribution: Understanding Cross-Entropy Ranking in Information Retrieval

DTIC Science & Technology

2006-07-01

reflect those of the spon- sor. viii ABSTRACT Unigram Language modeling is a successful probabilistic framework for Information Retrieval (IR) that uses...the Relevance model (RM), a state-of-the-art model for IR in the language modeling framework that uses the same cross-entropy as its ranking function...In addition, the SD based classifier provides more flexibility than RM in modeling documents owing to a consistent generative framework . We

Development of a Probabilistic Decision-Support Model to Forecast Coastal Resilience

NASA Astrophysics Data System (ADS)

Wilson, K.; Safak, I.; Brenner, O.; Lentz, E. E.; Hapke, C. J.

2016-02-01

Site-specific forecasts of coastal change are a valuable management tool in preparing for and assessing storm-driven impacts in coastal areas. More specifically, understanding the likelihood of storm impacts, recovery following events, and the alongshore variability of both is central in evaluating vulnerability and resiliency of barrier islands. We introduce a probabilistic modeling framework that integrates hydrodynamic, anthropogenic, and morphologic components of the barrier system to evaluate coastal change at Fire Island, New York. The model is structured on a Bayesian network (BN), which utilizes observations to learn statistical relationships between system variables. In addition to predictive ability, probabilistic models convey the level of confidence associated with a prediction, an important consideration for coastal managers. Our model predicts the likelihood of morphologic change on the upper beach based on several decades of beach monitoring data. A coupled hydrodynamic BN combines probabilistic and deterministic modeling approaches; by querying nearly two decades of nested-grid wave simulations that account for both distant swells and local seas, we produce scenarios of event and seasonal wave climates. The wave scenarios of total water level - a sum of run up, surge and tide - and anthropogenic modification are the primary drivers of morphologic change in our model structure. Preliminary results show the hydrodynamic BN is able to reproduce time series of total water levels, a critical validation process before generating scenarios, and forecasts of geomorphic change over three month intervals are up to 70% accurate. Predictions of storm-induced change and recovery are linked to evaluate zones of persistent vulnerability or resilience and will help managers target restoration efforts, identify areas most vulnerable to habitat degradation, and highlight resilient zones that may best support relocation of critical infrastructure.

Asymmetric author-topic model for knowledge discovering of big data in toxicogenomics.

PubMed

Chung, Ming-Hua; Wang, Yuping; Tang, Hailin; Zou, Wen; Basinger, John; Xu, Xiaowei; Tong, Weida

2015-01-01

The advancement of high-throughput screening technologies facilitates the generation of massive amount of biological data, a big data phenomena in biomedical science. Yet, researchers still heavily rely on keyword search and/or literature review to navigate the databases and analyses are often done in rather small-scale. As a result, the rich information of a database has not been fully utilized, particularly for the information embedded in the interactive nature between data points that are largely ignored and buried. For the past 10 years, probabilistic topic modeling has been recognized as an effective machine learning algorithm to annotate the hidden thematic structure of massive collection of documents. The analogy between text corpus and large-scale genomic data enables the application of text mining tools, like probabilistic topic models, to explore hidden patterns of genomic data and to the extension of altered biological functions. In this paper, we developed a generalized probabilistic topic model to analyze a toxicogenomics dataset that consists of a large number of gene expression data from the rat livers treated with drugs in multiple dose and time-points. We discovered the hidden patterns in gene expression associated with the effect of doses and time-points of treatment. Finally, we illustrated the ability of our model to identify the evidence of potential reduction of animal use.

Landslide Hazard from Coupled Inherent and Dynamic Probabilities

NASA Astrophysics Data System (ADS)

Strauch, R. L.; Istanbulluoglu, E.; Nudurupati, S. S.

2015-12-01

Landslide hazard research has typically been conducted independently from hydroclimate research. We sought to unify these two lines of research to provide regional scale landslide hazard information for risk assessments and resource management decision-making. Our approach couples an empirical inherent landslide probability, based on a frequency ratio analysis, with a numerical dynamic probability, generated by combining subsurface water recharge and surface runoff from the Variable Infiltration Capacity (VIC) macro-scale land surface hydrologic model with a finer resolution probabilistic slope stability model. Landslide hazard mapping is advanced by combining static and dynamic models of stability into a probabilistic measure of geohazard prediction in both space and time. This work will aid resource management decision-making in current and future landscape and climatic conditions. The approach is applied as a case study in North Cascade National Park Complex in northern Washington State.

3D Traffic Scene Understanding From Movable Platforms.

PubMed

Geiger, Andreas; Lauer, Martin; Wojek, Christian; Stiller, Christoph; Urtasun, Raquel

2014-05-01

In this paper, we present a novel probabilistic generative model for multi-object traffic scene understanding from movable platforms which reasons jointly about the 3D scene layout as well as the location and orientation of objects in the scene. In particular, the scene topology, geometry, and traffic activities are inferred from short video sequences. Inspired by the impressive driving capabilities of humans, our model does not rely on GPS, lidar, or map knowledge. Instead, it takes advantage of a diverse set of visual cues in the form of vehicle tracklets, vanishing points, semantic scene labels, scene flow, and occupancy grids. For each of these cues, we propose likelihood functions that are integrated into a probabilistic generative model. We learn all model parameters from training data using contrastive divergence. Experiments conducted on videos of 113 representative intersections show that our approach successfully infers the correct layout in a variety of very challenging scenarios. To evaluate the importance of each feature cue, experiments using different feature combinations are conducted. Furthermore, we show how by employing context derived from the proposed method we are able to improve over the state-of-the-art in terms of object detection and object orientation estimation in challenging and cluttered urban environments.

Bayesian accounts of covert selective attention: A tutorial review.

PubMed

Vincent, Benjamin T

2015-05-01

Decision making and optimal observer models offer an important theoretical approach to the study of covert selective attention. While their probabilistic formulation allows quantitative comparison to human performance, the models can be complex and their insights are not always immediately apparent. Part 1 establishes the theoretical appeal of the Bayesian approach, and introduces the way in which probabilistic approaches can be applied to covert search paradigms. Part 2 presents novel formulations of Bayesian models of 4 important covert attention paradigms, illustrating optimal observer predictions over a range of experimental manipulations. Graphical model notation is used to present models in an accessible way and Supplementary Code is provided to help bridge the gap between model theory and practical implementation. Part 3 reviews a large body of empirical and modelling evidence showing that many experimental phenomena in the domain of covert selective attention are a set of by-products. These effects emerge as the result of observers conducting Bayesian inference with noisy sensory observations, prior expectations, and knowledge of the generative structure of the stimulus environment.

Ensemble Bayesian forecasting system Part I: Theory and algorithms

NASA Astrophysics Data System (ADS)

Herr, Henry D.; Krzysztofowicz, Roman

2015-05-01

The ensemble Bayesian forecasting system (EBFS), whose theory was published in 2001, is developed for the purpose of quantifying the total uncertainty about a discrete-time, continuous-state, non-stationary stochastic process such as a time series of stages, discharges, or volumes at a river gauge. The EBFS is built of three components: an input ensemble forecaster (IEF), which simulates the uncertainty associated with random inputs; a deterministic hydrologic model (of any complexity), which simulates physical processes within a river basin; and a hydrologic uncertainty processor (HUP), which simulates the hydrologic uncertainty (an aggregate of all uncertainties except input). It works as a Monte Carlo simulator: an ensemble of time series of inputs (e.g., precipitation amounts) generated by the IEF is transformed deterministically through a hydrologic model into an ensemble of time series of outputs, which is next transformed stochastically by the HUP into an ensemble of time series of predictands (e.g., river stages). Previous research indicated that in order to attain an acceptable sampling error, the ensemble size must be on the order of hundreds (for probabilistic river stage forecasts and probabilistic flood forecasts) or even thousands (for probabilistic stage transition forecasts). The computing time needed to run the hydrologic model this many times renders the straightforward simulations operationally infeasible. This motivates the development of the ensemble Bayesian forecasting system with randomization (EBFSR), which takes full advantage of the analytic meta-Gaussian HUP and generates multiple ensemble members after each run of the hydrologic model; this auxiliary randomization reduces the required size of the meteorological input ensemble and makes it operationally feasible to generate a Bayesian ensemble forecast of large size. Such a forecast quantifies the total uncertainty, is well calibrated against the prior (climatic) distribution of predictand, possesses a Bayesian coherence property, constitutes a random sample of the predictand, and has an acceptable sampling error-which makes it suitable for rational decision making under uncertainty.

Probabilistic forecasts based on radar rainfall uncertainty

NASA Astrophysics Data System (ADS)

Liguori, S.; Rico-Ramirez, M. A.

2012-04-01

The potential advantages resulting from integrating weather radar rainfall estimates in hydro-meteorological forecasting systems is limited by the inherent uncertainty affecting radar rainfall measurements, which is due to various sources of error [1-3]. The improvement of quality control and correction techniques is recognized to play a role for the future improvement of radar-based flow predictions. However, the knowledge of the uncertainty affecting radar rainfall data can also be effectively used to build a hydro-meteorological forecasting system in a probabilistic framework. This work discusses the results of the implementation of a novel probabilistic forecasting system developed to improve ensemble predictions over a small urban area located in the North of England. An ensemble of radar rainfall fields can be determined as the sum of a deterministic component and a perturbation field, the latter being informed by the knowledge of the spatial-temporal characteristics of the radar error assessed with reference to rain-gauges measurements. This approach is similar to the REAL system [4] developed for use in the Southern-Alps. The radar uncertainty estimate can then be propagated with a nowcasting model, used to extrapolate an ensemble of radar rainfall forecasts, which can ultimately drive hydrological ensemble predictions. A radar ensemble generator has been calibrated using radar rainfall data made available from the UK Met Office after applying post-processing and corrections algorithms [5-6]. One hour rainfall accumulations from 235 rain gauges recorded for the year 2007 have provided the reference to determine the radar error. Statistics describing the spatial characteristics of the error (i.e. mean and covariance) have been computed off-line at gauges location, along with the parameters describing the error temporal correlation. A system has then been set up to impose the space-time error properties to stochastic perturbations, generated in real-time at gauges location, and then interpolated back onto the radar domain, in order to obtain probabilistic radar rainfall fields in real time. The deterministic nowcasting model integrated in the STEPS system [7-8] has been used for the purpose of propagating the uncertainty and assessing the benefit of implementing the radar ensemble generator for probabilistic rainfall forecasts and ultimately sewer flow predictions. For this purpose, events representative of different types of precipitation (i.e. stratiform/convective) and significant at the urban catchment scale (i.e. in terms of sewer overflow within the urban drainage system) have been selected. As high spatial/temporal resolution is required to the forecasts for their use in urban areas [9-11], the probabilistic nowcasts have been set up to be produced at 1 km resolution and 5 min intervals. The forecasting chain is completed by a hydrodynamic model of the urban drainage network. The aim of this work is to discuss the implementation of this probabilistic system, which takes into account the radar error to characterize the forecast uncertainty, with consequent potential benefits in the management of urban systems. It will also allow a comparison with previous findings related to the analysis of different approaches to uncertainty estimation and quantification in terms of rainfall [12] and flows at the urban scale [13]. Acknowledgements The authors would like to acknowledge the BADC, the UK Met Office and Dr. Alan Seed from the Australian Bureau of Meteorology for providing the radar data and the nowcasting model. The authors acknowledge the support from the Engineering and Physical Sciences Research Council (EPSRC) via grant EP/I012222/1.

Monte Carlo Approach for Reliability Estimations in Generalizability Studies.

ERIC Educational Resources Information Center

Dimitrov, Dimiter M.

A Monte Carlo approach is proposed, using the Statistical Analysis System (SAS) programming language, for estimating reliability coefficients in generalizability theory studies. Test scores are generated by a probabilistic model that considers the probability for a person with a given ability score to answer an item with a given difficulty…

Seismically induced landslides: current research by the US Geological Survey.

USGS Publications Warehouse

Harp, E.L.; Wilson, R.C.; Keefer, D.K.; Wieczorek, G.F.

1986-01-01

We have produced a regional seismic slope-stability map and a probabilistic prediction of landslide distribution from a postulated earthquake. For liquefaction-induced landslides, in situ measurements of seismically induced pore-water pressures have been used to establish an elastic model of pore pressure generation. -from Authors

PBDE exposure from food in Ireland: optimising data exploitation in probabilistic exposure modelling.

PubMed

Trudel, David; Tlustos, Christina; Von Goetz, Natalie; Scheringer, Martin; Hungerbühler, Konrad

2011-01-01

Polybrominated diphenyl ethers (PBDEs) are a class of brominated flame retardants added to plastics, polyurethane foam, electronics, textiles, and other products. These products release PBDEs into the indoor and outdoor environment, thus causing human exposure through food and dust. This study models PBDE dose distributions from ingestion of food for Irish adults on congener basis by using two probabilistic and one semi-deterministic method. One of the probabilistic methods was newly developed and is based on summary statistics of food consumption combined with a model generating realistic daily energy supply from food. Median (intermediate) doses of total PBDEs are in the range of 0.4-0.6 ng/kg(bw)/day for Irish adults. The 97.5th percentiles of total PBDE doses lie in a range of 1.7-2.2 ng/kg(bw)/day, which is comparable to doses derived for Belgian and Dutch adults. BDE-47 and BDE-99 were identified as the congeners contributing most to estimated intakes, accounting for more than half of the total doses. The most influential food groups contributing to this intake are lean fish and salmon which together account for about 22-25% of the total doses.

Probabilistic stability analysis: the way forward for stability analysis of sustainable power systems.

PubMed

Milanović, Jovica V

2017-08-13

Future power systems will be significantly different compared with their present states. They will be characterized by an unprecedented mix of a wide range of electricity generation and transmission technologies, as well as responsive and highly flexible demand and storage devices with significant temporal and spatial uncertainty. The importance of probabilistic approaches towards power system stability analysis, as a subsection of power system studies routinely carried out by power system operators, has been highlighted in previous research. However, it may not be feasible (or even possible) to accurately model all of the uncertainties that exist within a power system. This paper describes for the first time an integral approach to probabilistic stability analysis of power systems, including small and large angular stability and frequency stability. It provides guidance for handling uncertainties in power system stability studies and some illustrative examples of the most recent results of probabilistic stability analysis of uncertain power systems.This article is part of the themed issue 'Energy management: flexibility, risk and optimization'. © 2017 The Author(s).

Probabilistic Analysis of Large-Scale Composite Structures Using the IPACS Code

NASA Technical Reports Server (NTRS)

Lemonds, Jeffrey; Kumar, Virendra

1995-01-01

An investigation was performed to ascertain the feasibility of using IPACS (Integrated Probabilistic Assessment of Composite Structures) for probabilistic analysis of a composite fan blade, the development of which is being pursued by various industries for the next generation of aircraft engines. A model representative of the class of fan blades used in the GE90 engine has been chosen as the structural component to be analyzed with IPACS. In this study, typical uncertainties are assumed in the level, and structural responses for ply stresses and frequencies are evaluated in the form of cumulative probability density functions. Because of the geometric complexity of the blade, the number of plies varies from several hundred at the root to about a hundred at the tip. This represents a extremely complex composites application for the IPACS code. A sensitivity study with respect to various random variables is also performed.

A probabilistic analysis of the crystal oscillator behavior at low drive levels

NASA Astrophysics Data System (ADS)

Shmaliy, Yuriy S.; Brendel, Rémi

2008-03-01

The paper discusses a probabilistic model of a crystal oscillator at low drive levels where the noise intensity is comparable with the oscillation amplitude. The stationary probability density of the oscillations envelope is derived and investigated for the nonlinear resonator loses. A stochastic explanation is given for the well-known phenomenon termed sleeping sickness associated with losing a facility of self-excitation by a crystal oscillator after a long storage without a power supply. It is shown that, with low drive levels leading to an insufficient feedback, a crystal oscillator generates the noise-induced oscillations rather than it absolutely "falls in sleep".

Variational Bayesian Inversion of Quasi-Localized Seismic Attributes for the Spatial Distribution of Geological Facies

NASA Astrophysics Data System (ADS)

Nawaz, Muhammad Atif; Curtis, Andrew

2018-04-01

We introduce a new Bayesian inversion method that estimates the spatial distribution of geological facies from attributes of seismic data, by showing how the usual probabilistic inverse problem can be solved using an optimization framework still providing full probabilistic results. Our mathematical model consists of seismic attributes as observed data, which are assumed to have been generated by the geological facies. The method infers the post-inversion (posterior) probability density of the facies plus some other unknown model parameters, from the seismic attributes and geological prior information. Most previous research in this domain is based on the localized likelihoods assumption, whereby the seismic attributes at a location are assumed to depend on the facies only at that location. Such an assumption is unrealistic because of imperfect seismic data acquisition and processing, and fundamental limitations of seismic imaging methods. In this paper, we relax this assumption: we allow probabilistic dependence between seismic attributes at a location and the facies in any neighbourhood of that location through a spatial filter. We term such likelihoods quasi-localized.

Supernova Cosmology Inference with Probabilistic Photometric Redshifts (SCIPPR)

NASA Astrophysics Data System (ADS)

Peters, Christina; Malz, Alex; Hlozek, Renée

2018-01-01

The Bayesian Estimation Applied to Multiple Species (BEAMS) framework employs probabilistic supernova type classifications to do photometric SN cosmology. This work extends BEAMS to replace high-confidence spectroscopic redshifts with photometric redshift probability density functions, a capability that will be essential in the era the Large Synoptic Survey Telescope and other next-generation photometric surveys where it will not be possible to perform spectroscopic follow up on every SN. We present the Supernova Cosmology Inference with Probabilistic Photometric Redshifts (SCIPPR) Bayesian hierarchical model for constraining the cosmological parameters from photometric lightcurves and host galaxy photometry, which includes selection effects and is extensible to uncertainty in the redshift-dependent supernova type proportions. We create a pair of realistic mock catalogs of joint posteriors over supernova type, redshift, and distance modulus informed by photometric supernova lightcurves and over redshift from simulated host galaxy photometry. We perform inference under our model to obtain a joint posterior probability distribution over the cosmological parameters and compare our results with other methods, namely: a spectroscopic subset, a subset of high probability photometrically classified supernovae, and reducing the photometric redshift probability to a single measurement and error bar.

Learning Probabilistic Logic Models from Probabilistic Examples

PubMed Central

Chen, Jianzhong; Muggleton, Stephen; Santos, José

2009-01-01

Abstract We revisit an application developed originally using abductive Inductive Logic Programming (ILP) for modeling inhibition in metabolic networks. The example data was derived from studies of the effects of toxins on rats using Nuclear Magnetic Resonance (NMR) time-trace analysis of their biofluids together with background knowledge representing a subset of the Kyoto Encyclopedia of Genes and Genomes (KEGG). We now apply two Probabilistic ILP (PILP) approaches - abductive Stochastic Logic Programs (SLPs) and PRogramming In Statistical modeling (PRISM) to the application. Both approaches support abductive learning and probability predictions. Abductive SLPs are a PILP framework that provides possible worlds semantics to SLPs through abduction. Instead of learning logic models from non-probabilistic examples as done in ILP, the PILP approach applied in this paper is based on a general technique for introducing probability labels within a standard scientific experimental setting involving control and treated data. Our results demonstrate that the PILP approach provides a way of learning probabilistic logic models from probabilistic examples, and the PILP models learned from probabilistic examples lead to a significant decrease in error accompanied by improved insight from the learned results compared with the PILP models learned from non-probabilistic examples. PMID:19888348

Learning Probabilistic Logic Models from Probabilistic Examples.

PubMed

Chen, Jianzhong; Muggleton, Stephen; Santos, José

2008-10-01

We revisit an application developed originally using abductive Inductive Logic Programming (ILP) for modeling inhibition in metabolic networks. The example data was derived from studies of the effects of toxins on rats using Nuclear Magnetic Resonance (NMR) time-trace analysis of their biofluids together with background knowledge representing a subset of the Kyoto Encyclopedia of Genes and Genomes (KEGG). We now apply two Probabilistic ILP (PILP) approaches - abductive Stochastic Logic Programs (SLPs) and PRogramming In Statistical modeling (PRISM) to the application. Both approaches support abductive learning and probability predictions. Abductive SLPs are a PILP framework that provides possible worlds semantics to SLPs through abduction. Instead of learning logic models from non-probabilistic examples as done in ILP, the PILP approach applied in this paper is based on a general technique for introducing probability labels within a standard scientific experimental setting involving control and treated data. Our results demonstrate that the PILP approach provides a way of learning probabilistic logic models from probabilistic examples, and the PILP models learned from probabilistic examples lead to a significant decrease in error accompanied by improved insight from the learned results compared with the PILP models learned from non-probabilistic examples.

Integrated Risk-Informed Decision-Making for an ALMR PRISM

DOE Office of Scientific and Technical Information (OSTI.GOV)

Muhlheim, Michael David; Belles, Randy; Denning, Richard S.

Decision-making is the process of identifying decision alternatives, assessing those alternatives based on predefined metrics, selecting an alternative (i.e., making a decision), and then implementing that alternative. The generation of decisions requires a structured, coherent process, or a decision-making process. The overall objective for this work is that the generalized framework is adopted into an autonomous decision-making framework and tailored to specific requirements for various applications. In this context, automation is the use of computing resources to make decisions and implement a structured decision-making process with limited or no human intervention. The overriding goal of automation is to replace ormore » supplement human decision makers with reconfigurable decision-making modules that can perform a given set of tasks rationally, consistently, and reliably. Risk-informed decision-making requires a probabilistic assessment of the likelihood of success given the status of the plant/systems and component health, and a deterministic assessment between plant operating parameters and reactor protection parameters to prevent unnecessary trips and challenges to plant safety systems. The probabilistic portion of the decision-making engine of the supervisory control system is based on the control actions associated with an ALMR PRISM. Newly incorporated into the probabilistic models are the prognostic/diagnostic models developed by Pacific Northwest National Laboratory. These allow decisions to incorporate the health of components into the decision–making process. Once the control options are identified and ranked based on the likelihood of success, the supervisory control system transmits the options to the deterministic portion of the platform. The deterministic portion of the decision-making engine uses thermal-hydraulic modeling and components for an advanced liquid-metal reactor Power Reactor Inherently Safe Module. The deterministic multi-attribute decision-making framework uses various sensor data (e.g., reactor outlet temperature, steam generator drum level) and calculates its position within the challenge state, its trajectory, and its margin within the controllable domain using utility functions to evaluate current and projected plant state space for different control decisions. The metrics that are evaluated are based on reactor trip set points. The integration of the deterministic calculations using multi-physics analyses and probabilistic safety calculations allows for the examination and quantification of margin recovery strategies. This also provides validation of the control options identified from the probabilistic assessment. Thus, the thermalhydraulics analyses are used to validate the control options identified from the probabilistic assessment. Future work includes evaluating other possible metrics and computational efficiencies, and developing a user interface to mimic display panels at a modern nuclear power plant.« less

Evaluating and comparing methods of sinkhole susceptibility mapping in the Ebro Valley evaporite karst (NE Spain)

NASA Astrophysics Data System (ADS)

Galve, J. P.; Gutiérrez, F.; Remondo, J.; Bonachea, J.; Lucha, P.; Cendrero, A.

2009-10-01

Multiple sinkhole susceptibility models have been generated in three study areas of the Ebro Valley evaporite karst (NE Spain) applying different methods (nearest neighbour distance, sinkhole density, heuristic scoring system and probabilistic analysis) for each sinkhole type separately (cover collapse sinkholes, cover and bedrock collapse sinkholes and cover and bedrock sagging sinkholes). The quantitative and independent evaluation of the predictive capability of the models reveals that: (1) The most reliable susceptibility models are those derived from the nearest neighbour distance and sinkhole density. These models can be generated in a simple and rapid way from detailed geomorphological maps. (2) The reliability of the nearest neighbour distance and density models is conditioned by the degree of clustering of the sinkholes. Consequently, the karst areas in which sinkholes show a higher clustering are a priori more favourable for predicting new occurrences. (3) The predictive capability of the best models obtained in this research is significantly higher (12.5-82.5%) than that of the heuristic sinkhole susceptibility model incorporated into the General Urban Plan for the municipality of Zaragoza. Although the probabilistic approach provides lower quality results than the methods based on sinkhole proximity and density, it helps to identify the most significant factors and select the most effective mitigation strategies and may be applied to model susceptibility in different future scenarios.

SU-E-I-87: Automated Liver Segmentation Method for CBCT Dataset by Combining Sparse Shape Composition and Probabilistic Atlas Construction

DOE Office of Scientific and Technical Information (OSTI.GOV)

Li, Dengwang; Liu, Li; Chen, Jinhu

2014-06-01

Purpose: The aiming of this study was to extract liver structures for daily Cone beam CT (CBCT) images automatically. Methods: Datasets were collected from 50 intravenous contrast planning CT images, which were regarded as training dataset for probabilistic atlas and shape prior model construction. Firstly, probabilistic atlas and shape prior model based on sparse shape composition (SSC) were constructed by iterative deformable registration. Secondly, the artifacts and noise were removed from the daily CBCT image by an edge-preserving filtering using total variation with L1 norm (TV-L1). Furthermore, the initial liver region was obtained by registering the incoming CBCT image withmore » the atlas utilizing edge-preserving deformable registration with multi-scale strategy, and then the initial liver region was converted to surface meshing which was registered with the shape model where the major variation of specific patient was modeled by sparse vectors. At the last stage, the shape and intensity information were incorporated into joint probabilistic model, and finally the liver structure was extracted by maximum a posteriori segmentation.Regarding the construction process, firstly the manually segmented contours were converted into meshes, and then arbitrary patient data was chosen as reference image to register with the rest of training datasets by deformable registration algorithm for constructing probabilistic atlas and prior shape model. To improve the efficiency of proposed method, the initial probabilistic atlas was used as reference image to register with other patient data for iterative construction for removing bias caused by arbitrary selection. Results: The experiment validated the accuracy of the segmentation results quantitatively by comparing with the manually ones. The volumetric overlap percentage between the automatically generated liver contours and the ground truth were on an average 88%–95% for CBCT images. Conclusion: The experiment demonstrated that liver structures of CBCT with artifacts can be extracted accurately for following adaptive radiation therapy. This work is supported by National Natural Science Foundation of China (No. 61201441), Research Fund for Excellent Young and Middle-aged Scientists of Shandong Province (No. BS2012DX038), Project of Shandong Province Higher Educational Science and Technology Program (No. J12LN23), Jinan youth science and technology star (No.20120109)« less

Model-based machine learning.

PubMed

Bishop, Christopher M

2013-02-13

Several decades of research in the field of machine learning have resulted in a multitude of different algorithms for solving a broad range of problems. To tackle a new application, a researcher typically tries to map their problem onto one of these existing methods, often influenced by their familiarity with specific algorithms and by the availability of corresponding software implementations. In this study, we describe an alternative methodology for applying machine learning, in which a bespoke solution is formulated for each new application. The solution is expressed through a compact modelling language, and the corresponding custom machine learning code is then generated automatically. This model-based approach offers several major advantages, including the opportunity to create highly tailored models for specific scenarios, as well as rapid prototyping and comparison of a range of alternative models. Furthermore, newcomers to the field of machine learning do not have to learn about the huge range of traditional methods, but instead can focus their attention on understanding a single modelling environment. In this study, we show how probabilistic graphical models, coupled with efficient inference algorithms, provide a very flexible foundation for model-based machine learning, and we outline a large-scale commercial application of this framework involving tens of millions of users. We also describe the concept of probabilistic programming as a powerful software environment for model-based machine learning, and we discuss a specific probabilistic programming language called Infer.NET, which has been widely used in practical applications.

Model-based machine learning

PubMed Central

Bishop, Christopher M.

2013-01-01

Several decades of research in the field of machine learning have resulted in a multitude of different algorithms for solving a broad range of problems. To tackle a new application, a researcher typically tries to map their problem onto one of these existing methods, often influenced by their familiarity with specific algorithms and by the availability of corresponding software implementations. In this study, we describe an alternative methodology for applying machine learning, in which a bespoke solution is formulated for each new application. The solution is expressed through a compact modelling language, and the corresponding custom machine learning code is then generated automatically. This model-based approach offers several major advantages, including the opportunity to create highly tailored models for specific scenarios, as well as rapid prototyping and comparison of a range of alternative models. Furthermore, newcomers to the field of machine learning do not have to learn about the huge range of traditional methods, but instead can focus their attention on understanding a single modelling environment. In this study, we show how probabilistic graphical models, coupled with efficient inference algorithms, provide a very flexible foundation for model-based machine learning, and we outline a large-scale commercial application of this framework involving tens of millions of users. We also describe the concept of probabilistic programming as a powerful software environment for model-based machine learning, and we discuss a specific probabilistic programming language called Infer.NET, which has been widely used in practical applications. PMID:23277612

U.S. Patent Pending, Information Security Analysis Using Game Theory and Simulation, U.S. Patent Application No.: 14/097,840

DOE Office of Scientific and Technical Information (OSTI.GOV)

Abercrombie, Robert K; Schlicher, Bob G

Vulnerability in security of an information system is quantitatively predicted. The information system may receive malicious actions against its security and may receive corrective actions for restoring the security. A game oriented agent based model is constructed in a simulator application. The game ABM model represents security activity in the information system. The game ABM model has two opposing participants including an attacker and a defender, probabilistic game rules and allowable game states. A specified number of simulations are run and a probabilistic number of the plurality of allowable game states are reached in each simulation run. The probability ofmore » reaching a specified game state is unknown prior to running each simulation. Data generated during the game states is collected to determine a probability of one or more aspects of security in the information system.« less

A framework for the probabilistic analysis of meteotsunamis

USGS Publications Warehouse

Geist, Eric L.; ten Brink, Uri S.; Gove, Matthew D.

2014-01-01

A probabilistic technique is developed to assess the hazard from meteotsunamis. Meteotsunamis are unusual sea-level events, generated when the speed of an atmospheric pressure or wind disturbance is comparable to the phase speed of long waves in the ocean. A general aggregation equation is proposed for the probabilistic analysis, based on previous frameworks established for both tsunamis and storm surges, incorporating different sources and source parameters of meteotsunamis. Parameterization of atmospheric disturbances and numerical modeling is performed for the computation of maximum meteotsunami wave amplitudes near the coast. A historical record of pressure disturbances is used to establish a continuous analytic distribution of each parameter as well as the overall Poisson rate of occurrence. A demonstration study is presented for the northeast U.S. in which only isolated atmospheric pressure disturbances from squall lines and derechos are considered. For this study, Automated Surface Observing System stations are used to determine the historical parameters of squall lines from 2000 to 2013. The probabilistic equations are implemented using a Monte Carlo scheme, where a synthetic catalog of squall lines is compiled by sampling the parameter distributions. For each entry in the catalog, ocean wave amplitudes are computed using a numerical hydrodynamic model. Aggregation of the results from the Monte Carlo scheme results in a meteotsunami hazard curve that plots the annualized rate of exceedance with respect to maximum event amplitude for a particular location along the coast. Results from using multiple synthetic catalogs, resampled from the parent parameter distributions, yield mean and quantile hazard curves. Further refinements and improvements for probabilistic analysis of meteotsunamis are discussed.

A High Performance Computing Approach to Tree Cover Delineation in 1-m NAIP Imagery Using a Probabilistic Learning Framework

NASA Technical Reports Server (NTRS)

Basu, Saikat; Ganguly, Sangram; Michaelis, Andrew; Votava, Petr; Roy, Anshuman; Mukhopadhyay, Supratik; Nemani, Ramakrishna

2015-01-01

Tree cover delineation is a useful instrument in deriving Above Ground Biomass (AGB) density estimates from Very High Resolution (VHR) airborne imagery data. Numerous algorithms have been designed to address this problem, but most of them do not scale to these datasets, which are of the order of terabytes. In this paper, we present a semi-automated probabilistic framework for the segmentation and classification of 1-m National Agriculture Imagery Program (NAIP) for tree-cover delineation for the whole of Continental United States, using a High Performance Computing Architecture. Classification is performed using a multi-layer Feedforward Backpropagation Neural Network and segmentation is performed using a Statistical Region Merging algorithm. The results from the classification and segmentation algorithms are then consolidated into a structured prediction framework using a discriminative undirected probabilistic graphical model based on Conditional Random Field, which helps in capturing the higher order contextual dependencies between neighboring pixels. Once the final probability maps are generated, the framework is updated and re-trained by relabeling misclassified image patches. This leads to a significant improvement in the true positive rates and reduction in false positive rates. The tree cover maps were generated for the whole state of California, spanning a total of 11,095 NAIP tiles covering a total geographical area of 163,696 sq. miles. The framework produced true positive rates of around 88% for fragmented forests and 74% for urban tree cover areas, with false positive rates lower than 2% for both landscapes. Comparative studies with the National Land Cover Data (NLCD) algorithm and the LiDAR canopy height model (CHM) showed the effectiveness of our framework for generating accurate high-resolution tree-cover maps.

A High Performance Computing Approach to Tree Cover Delineation in 1-m NAIP Imagery using a Probabilistic Learning Framework

NASA Astrophysics Data System (ADS)

Basu, S.; Ganguly, S.; Michaelis, A.; Votava, P.; Roy, A.; Mukhopadhyay, S.; Nemani, R. R.

2015-12-01

Tree cover delineation is a useful instrument in deriving Above Ground Biomass (AGB) density estimates from Very High Resolution (VHR) airborne imagery data. Numerous algorithms have been designed to address this problem, but most of them do not scale to these datasets which are of the order of terabytes. In this paper, we present a semi-automated probabilistic framework for the segmentation and classification of 1-m National Agriculture Imagery Program (NAIP) for tree-cover delineation for the whole of Continental United States, using a High Performance Computing Architecture. Classification is performed using a multi-layer Feedforward Backpropagation Neural Network and segmentation is performed using a Statistical Region Merging algorithm. The results from the classification and segmentation algorithms are then consolidated into a structured prediction framework using a discriminative undirected probabilistic graphical model based on Conditional Random Field, which helps in capturing the higher order contextual dependencies between neighboring pixels. Once the final probability maps are generated, the framework is updated and re-trained by relabeling misclassified image patches. This leads to a significant improvement in the true positive rates and reduction in false positive rates. The tree cover maps were generated for the whole state of California, spanning a total of 11,095 NAIP tiles covering a total geographical area of 163,696 sq. miles. The framework produced true positive rates of around 88% for fragmented forests and 74% for urban tree cover areas, with false positive rates lower than 2% for both landscapes. Comparative studies with the National Land Cover Data (NLCD) algorithm and the LiDAR canopy height model (CHM) showed the effectiveness of our framework for generating accurate high-resolution tree-cover maps.

A Proposed Probabilistic Extension of the Halpern and Pearl Definition of ‘Actual Cause’

PubMed Central

2017-01-01

ABSTRACT Joseph Halpern and Judea Pearl ([2005]) draw upon structural equation models to develop an attractive analysis of ‘actual cause’. Their analysis is designed for the case of deterministic causation. I show that their account can be naturally extended to provide an elegant treatment of probabilistic causation. 1Introduction2Preemption3Structural Equation Models4The Halpern and Pearl Definition of ‘Actual Cause’5Preemption Again6The Probabilistic Case7Probabilistic Causal Models8A Proposed Probabilistic Extension of Halpern and Pearl’s Definition9Twardy and Korb’s Account10Probabilistic Fizzling11Conclusion PMID:29593362

Probabilistic generation of random networks taking into account information on motifs occurrence.

PubMed

Bois, Frederic Y; Gayraud, Ghislaine

2015-01-01

Because of the huge number of graphs possible even with a small number of nodes, inference on network structure is known to be a challenging problem. Generating large random directed graphs with prescribed probabilities of occurrences of some meaningful patterns (motifs) is also difficult. We show how to generate such random graphs according to a formal probabilistic representation, using fast Markov chain Monte Carlo methods to sample them. As an illustration, we generate realistic graphs with several hundred nodes mimicking a gene transcription interaction network in Escherichia coli.

Probabilistic Generation of Random Networks Taking into Account Information on Motifs Occurrence

PubMed Central

Bois, Frederic Y.

2015-01-01

Abstract Because of the huge number of graphs possible even with a small number of nodes, inference on network structure is known to be a challenging problem. Generating large random directed graphs with prescribed probabilities of occurrences of some meaningful patterns (motifs) is also difficult. We show how to generate such random graphs according to a formal probabilistic representation, using fast Markov chain Monte Carlo methods to sample them. As an illustration, we generate realistic graphs with several hundred nodes mimicking a gene transcription interaction network in Escherichia coli. PMID:25493547

Probabilistic tsunami hazard assessment at Seaside, Oregon, for near-and far-field seismic sources

USGS Publications Warehouse

Gonzalez, F.I.; Geist, E.L.; Jaffe, B.; Kanoglu, U.; Mofjeld, H.; Synolakis, C.E.; Titov, V.V.; Areas, D.; Bellomo, D.; Carlton, D.; Horning, T.; Johnson, J.; Newman, J.; Parsons, T.; Peters, R.; Peterson, C.; Priest, G.; Venturato, A.; Weber, J.; Wong, F.; Yalciner, A.

2009-01-01

The first probabilistic tsunami flooding maps have been developed. The methodology, called probabilistic tsunami hazard assessment (PTHA), integrates tsunami inundation modeling with methods of probabilistic seismic hazard assessment (PSHA). Application of the methodology to Seaside, Oregon, has yielded estimates of the spatial distribution of 100- and 500-year maximum tsunami amplitudes, i.e., amplitudes with 1% and 0.2% annual probability of exceedance. The 100-year tsunami is generated most frequently by far-field sources in the Alaska-Aleutian Subduction Zone and is characterized by maximum amplitudes that do not exceed 4 m, with an inland extent of less than 500 m. In contrast, the 500-year tsunami is dominated by local sources in the Cascadia Subduction Zone and is characterized by maximum amplitudes in excess of 10 m and an inland extent of more than 1 km. The primary sources of uncertainty in these results include those associated with interevent time estimates, modeling of background sea level, and accounting for temporal changes in bathymetry and topography. Nonetheless, PTHA represents an important contribution to tsunami hazard assessment techniques; viewed in the broader context of risk analysis, PTHA provides a method for quantifying estimates of the likelihood and severity of the tsunami hazard, which can then be combined with vulnerability and exposure to yield estimates of tsunami risk. Copyright 2009 by the American Geophysical Union.

An Improved, Bias-Reduced Probabilistic Functional Gene Network of Baker's Yeast, Saccharomyces cerevisiae

PubMed Central

Lee, Insuk; Li, Zhihua; Marcotte, Edward M.

2007-01-01

Background Probabilistic functional gene networks are powerful theoretical frameworks for integrating heterogeneous functional genomics and proteomics data into objective models of cellular systems. Such networks provide syntheses of millions of discrete experimental observations, spanning DNA microarray experiments, physical protein interactions, genetic interactions, and comparative genomics; the resulting networks can then be easily applied to generate testable hypotheses regarding specific gene functions and associations. Methodology/Principal Findings We report a significantly improved version (v. 2) of a probabilistic functional gene network [1] of the baker's yeast, Saccharomyces cerevisiae. We describe our optimization methods and illustrate their effects in three major areas: the reduction of functional bias in network training reference sets, the application of a probabilistic model for calculating confidences in pair-wise protein physical or genetic interactions, and the introduction of simple thresholds that eliminate many false positive mRNA co-expression relationships. Using the network, we predict and experimentally verify the function of the yeast RNA binding protein Puf6 in 60S ribosomal subunit biogenesis. Conclusions/Significance YeastNet v. 2, constructed using these optimizations together with additional data, shows significant reduction in bias and improvements in precision and recall, in total covering 102,803 linkages among 5,483 yeast proteins (95% of the validated proteome). YeastNet is available from http://www.yeastnet.org. PMID:17912365

Probabilistic flood inundation mapping at ungauged streams due to roughness coefficient uncertainty in hydraulic modelling

NASA Astrophysics Data System (ADS)

Papaioannou, George; Vasiliades, Lampros; Loukas, Athanasios; Aronica, Giuseppe T.

2017-04-01

Probabilistic flood inundation mapping is performed and analysed at the ungauged Xerias stream reach, Volos, Greece. The study evaluates the uncertainty introduced by the roughness coefficient values on hydraulic models in flood inundation modelling and mapping. The well-established one-dimensional (1-D) hydraulic model, HEC-RAS is selected and linked to Monte-Carlo simulations of hydraulic roughness. Terrestrial Laser Scanner data have been used to produce a high quality DEM for input data uncertainty minimisation and to improve determination accuracy on stream channel topography required by the hydraulic model. Initial Manning's n roughness coefficient values are based on pebble count field surveys and empirical formulas. Various theoretical probability distributions are fitted and evaluated on their accuracy to represent the estimated roughness values. Finally, Latin Hypercube Sampling has been used for generation of different sets of Manning roughness values and flood inundation probability maps have been created with the use of Monte Carlo simulations. Historical flood extent data, from an extreme historical flash flood event, are used for validation of the method. The calibration process is based on a binary wet-dry reasoning with the use of Median Absolute Percentage Error evaluation metric. The results show that the proposed procedure supports probabilistic flood hazard mapping at ungauged rivers and provides water resources managers with valuable information for planning and implementing flood risk mitigation strategies.

Probabilistic Harmonic Analysis on Distributed Photovoltaic Integration Considering Typical Weather Scenarios

NASA Astrophysics Data System (ADS)

Bin, Che; Ruoying, Yu; Dongsheng, Dang; Xiangyan, Wang

2017-05-01

Distributed Generation (DG) integrating to the network would cause the harmonic pollution which would cause damages on electrical devices and affect the normal operation of power system. On the other hand, due to the randomness of the wind and solar irradiation, the output of DG is random, too, which leads to an uncertainty of the harmonic generated by the DG. Thus, probabilistic methods are needed to analyse the impacts of the DG integration. In this work we studied the harmonic voltage probabilistic distribution and the harmonic distortion in distributed network after the distributed photovoltaic (DPV) system integrating in different weather conditions, mainly the sunny day, cloudy day, rainy day and the snowy day. The probabilistic distribution function of the DPV output power in different typical weather conditions could be acquired via the parameter identification method of maximum likelihood estimation. The Monte-Carlo simulation method was adopted to calculate the probabilistic distribution of harmonic voltage content at different frequency orders as well as the harmonic distortion (THD) in typical weather conditions. The case study was based on the IEEE33 system and the results of harmonic voltage content probabilistic distribution as well as THD in typical weather conditions were compared.

Stochastic model simulation using Kronecker product analysis and Zassenhaus formula approximation.

PubMed

Caglar, Mehmet Umut; Pal, Ranadip

2013-01-01

Probabilistic Models are regularly applied in Genetic Regulatory Network modeling to capture the stochastic behavior observed in the generation of biological entities such as mRNA or proteins. Several approaches including Stochastic Master Equations and Probabilistic Boolean Networks have been proposed to model the stochastic behavior in genetic regulatory networks. It is generally accepted that Stochastic Master Equation is a fundamental model that can describe the system being investigated in fine detail, but the application of this model is computationally enormously expensive. On the other hand, Probabilistic Boolean Network captures only the coarse-scale stochastic properties of the system without modeling the detailed interactions. We propose a new approximation of the stochastic master equation model that is able to capture the finer details of the modeled system including bistabilities and oscillatory behavior, and yet has a significantly lower computational complexity. In this new method, we represent the system using tensors and derive an identity to exploit the sparse connectivity of regulatory targets for complexity reduction. The algorithm involves an approximation based on Zassenhaus formula to represent the exponential of a sum of matrices as product of matrices. We derive upper bounds on the expected error of the proposed model distribution as compared to the stochastic master equation model distribution. Simulation results of the application of the model to four different biological benchmark systems illustrate performance comparable to detailed stochastic master equation models but with considerably lower computational complexity. The results also demonstrate the reduced complexity of the new approach as compared to commonly used Stochastic Simulation Algorithm for equivalent accuracy.

A temporal-spatial postprocessing model for probabilistic run-off forecast. With a case study from Ulla-Førre with five catchments and ten lead times

NASA Astrophysics Data System (ADS)

Engeland, K.; Steinsland, I.

2012-04-01

This work is driven by the needs of next generation short term optimization methodology for hydro power production. Stochastic optimization are about to be introduced; i.e. optimizing when available resources (water) and utility (prices) are uncertain. In this paper we focus on the available resources, i.e. water, where uncertainty mainly comes from uncertainty in future runoff. When optimizing a water system all catchments and several lead times have to be considered simultaneously. Depending on the system of hydropower reservoirs, it might be a set of headwater catchments, a system of upstream /downstream reservoirs where water used from one catchment /dam arrives in a lower catchment maybe days later, or a combination of both. The aim of this paper is therefore to construct a simultaneous probabilistic forecast for several catchments and lead times, i.e. to provide a predictive distribution for the forecasts. Stochastic optimization methods need samples/ensembles of run-off forecasts as input. Hence, it should also be possible to sample from our probabilistic forecast. A post-processing approach is taken, and an error model based on Box- Cox transformation, power transform and a temporal-spatial copula model is used. It accounts for both between catchment and between lead time dependencies. In operational use it is strait forward to sample run-off ensembles from this models that inherits the catchment and lead time dependencies. The methodology is tested and demonstrated in the Ulla-Førre river system, and simultaneous probabilistic forecasts for five catchments and ten lead times are constructed. The methodology has enough flexibility to model operationally important features in this case study such as hetroscadasety, lead-time varying temporal dependency and lead-time varying inter-catchment dependency. Our model is evaluated using CRPS for marginal predictive distributions and energy score for joint predictive distribution. It is tested against deterministic run-off forecast, climatology forecast and a persistent forecast, and is found to be the better probabilistic forecast for lead time grater then two. From an operational point of view the results are interesting as the between catchment dependency gets stronger with longer lead-times.

Dynamic Statistical Models for Pyroclastic Density Current Generation at Soufrière Hills Volcano

NASA Astrophysics Data System (ADS)

Wolpert, Robert L.; Spiller, Elaine T.; Calder, Eliza S.

2018-05-01

To mitigate volcanic hazards from pyroclastic density currents, volcanologists generate hazard maps that provide long-term forecasts of areas of potential impact. Several recent efforts in the field develop new statistical methods for application of flow models to generate fully probabilistic hazard maps that both account for, and quantify, uncertainty. However a limitation to the use of most statistical hazard models, and a key source of uncertainty within them, is the time-averaged nature of the datasets by which the volcanic activity is statistically characterized. Where the level, or directionality, of volcanic activity frequently changes, e.g. during protracted eruptive episodes, or at volcanoes that are classified as persistently active, it is not appropriate to make short term forecasts based on longer time-averaged metrics of the activity. Thus, here we build, fit and explore dynamic statistical models for the generation of pyroclastic density current from Soufrière Hills Volcano (SHV) on Montserrat including their respective collapse direction and flow volumes based on 1996-2008 flow datasets. The development of this approach allows for short-term behavioral changes to be taken into account in probabilistic volcanic hazard assessments. We show that collapses from the SHV lava dome follow a clear pattern, and that a series of smaller flows in a given direction often culminate in a larger collapse and thereafter directionality of the flows change. Such models enable short term forecasting (weeks to months) that can reflect evolving conditions such as dome and crater morphology changes and non-stationary eruptive behavior such as extrusion rate variations. For example, the probability of inundation of the Belham Valley in the first 180 days of a forecast period is about twice as high for lava domes facing Northwest toward that valley as it is for domes pointing East toward the Tar River Valley. As rich multi-parametric volcano monitoring dataset become increasingly available, eruption forecasting is becoming an increasingly viable and important research field. We demonstrate an approach to utilize such data in order to appropriately 'tune' probabilistic hazard assessments for pyroclastic flows. Our broader objective with development of this method is to help advance time-dependent volcanic hazard assessment, by bridging the

Bayesian Action-Perception loop modeling: Application to trajectory generation and recognition using internal motor simulation

NASA Astrophysics Data System (ADS)

Gilet, Estelle; Diard, Julien; Palluel-Germain, Richard; Bessière, Pierre

2011-03-01

This paper is about modeling perception-action loops and, more precisely, the study of the influence of motor knowledge during perception tasks. We use the Bayesian Action-Perception (BAP) model, which deals with the sensorimotor loop involved in reading and writing cursive isolated letters and includes an internal simulation of movement loop. By using this probabilistic model we simulate letter recognition, both with and without internal motor simulation. Comparison of their performance yields an experimental prediction, which we set forth.

Uncertainty Quantification of Evapotranspiration and Infiltration from Modeling and Historic Time Series at the Savannah River F-Area

NASA Astrophysics Data System (ADS)

Faybishenko, B.; Flach, G. P.

2012-12-01

The objectives of this presentation are: (a) to illustrate the application of Monte Carlo and fuzzy-probabilistic approaches for uncertainty quantification (UQ) in predictions of potential evapotranspiration (PET), actual evapotranspiration (ET), and infiltration (I), using uncertain hydrological or meteorological time series data, and (b) to compare the results of these calculations with those from field measurements at the U.S. Department of Energy Savannah River Site (SRS), near Aiken, South Carolina, USA. The UQ calculations include the evaluation of aleatory (parameter uncertainty) and epistemic (model) uncertainties. The effect of aleatory uncertainty is expressed by assigning the probability distributions of input parameters, using historical monthly averaged data from the meteorological station at the SRS. The combined effect of aleatory and epistemic uncertainties on the UQ of PET, ET, and Iis then expressed by aggregating the results of calculations from multiple models using a p-box and fuzzy numbers. The uncertainty in PETis calculated using the Bair-Robertson, Blaney-Criddle, Caprio, Hargreaves-Samani, Hamon, Jensen-Haise, Linacre, Makkink, Priestly-Taylor, Penman, Penman-Monteith, Thornthwaite, and Turc models. Then, ET is calculated from the modified Budyko model, followed by calculations of I from the water balance equation. We show that probabilistic and fuzzy-probabilistic calculations using multiple models generate the PET, ET, and Idistributions, which are well within the range of field measurements. We also show that a selection of a subset of models can be used to constrain the uncertainty quantification of PET, ET, and I.

Probabilistic Modeling of the Renal Stone Formation Module

NASA Technical Reports Server (NTRS)

Best, Lauren M.; Myers, Jerry G.; Goodenow, Debra A.; McRae, Michael P.; Jackson, Travis C.

2013-01-01

The Integrated Medical Model (IMM) is a probabilistic tool, used in mission planning decision making and medical systems risk assessments. The IMM project maintains a database of over 80 medical conditions that could occur during a spaceflight, documenting an incidence rate and end case scenarios for each. In some cases, where observational data are insufficient to adequately define the inflight medical risk, the IMM utilizes external probabilistic modules to model and estimate the event likelihoods. One such medical event of interest is an unpassed renal stone. Due to a high salt diet and high concentrations of calcium in the blood (due to bone depletion caused by unloading in the microgravity environment) astronauts are at a considerable elevated risk for developing renal calculi (nephrolithiasis) while in space. Lack of observed incidences of nephrolithiasis has led HRP to initiate the development of the Renal Stone Formation Module (RSFM) to create a probabilistic simulator capable of estimating the likelihood of symptomatic renal stone presentation in astronauts on exploration missions. The model consists of two major parts. The first is the probabilistic component, which utilizes probability distributions to assess the range of urine electrolyte parameters and a multivariate regression to transform estimated crystal density and size distributions to the likelihood of the presentation of nephrolithiasis symptoms. The second is a deterministic physical and chemical model of renal stone growth in the kidney developed by Kassemi et al. The probabilistic component of the renal stone model couples the input probability distributions describing the urine chemistry, astronaut physiology, and system parameters with the physical and chemical outputs and inputs to the deterministic stone growth model. These two parts of the model are necessary to capture the uncertainty in the likelihood estimate. The model will be driven by Monte Carlo simulations, continuously randomly sampling the probability distributions of the electrolyte concentrations and system parameters that are inputs into the deterministic model. The total urine chemistry concentrations are used to determine the urine chemistry activity using the Joint Expert Speciation System (JESS), a biochemistry model. Information used from JESS is then fed into the deterministic growth model. Outputs from JESS and the deterministic model are passed back to the probabilistic model where a multivariate regression is used to assess the likelihood of a stone forming and the likelihood of a stone requiring clinical intervention. The parameters used to determine to quantify these risks include: relative supersaturation (RS) of calcium oxalate, citrate/calcium ratio, crystal number density, total urine volume, pH, magnesium excretion, maximum stone width, and ureteral location. Methods and Validation: The RSFM is designed to perform a Monte Carlo simulation to generate probability distributions of clinically significant renal stones, as well as provide an associated uncertainty in the estimate. Initially, early versions will be used to test integration of the components and assess component validation and verification (V&V), with later versions used to address questions regarding design reference mission scenarios. Once integrated with the deterministic component, the credibility assessment of the integrated model will follow NASA STD 7009 requirements.

Evaluating the uncertainty of predicting future climate time series at the hourly time scale

NASA Astrophysics Data System (ADS)

Caporali, E.; Fatichi, S.; Ivanov, V. Y.

2011-12-01

A stochastic downscaling methodology is developed to generate hourly, point-scale time series for several meteorological variables, such as precipitation, cloud cover, shortwave radiation, air temperature, relative humidity, wind speed, and atmospheric pressure. The methodology uses multi-model General Circulation Model (GCM) realizations and an hourly weather generator, AWE-GEN. Probabilistic descriptions of factors of change (a measure of climate change with respect to historic conditions) are computed for several climate statistics and different aggregation times using a Bayesian approach that weights the individual GCM contributions. The Monte Carlo method is applied to sample the factors of change from their respective distributions thereby permitting the generation of time series in an ensemble fashion, which reflects the uncertainty of climate projections of future as well as the uncertainty of the downscaling procedure. Applications of the methodology and probabilistic expressions of certainty in reproducing future climates for the periods, 2000 - 2009, 2046 - 2065 and 2081 - 2100, using the 1962 - 1992 period as the baseline, are discussed for the location of Firenze (Italy). The climate predictions for the period of 2000 - 2009 are tested against observations permitting to assess the reliability and uncertainties of the methodology in reproducing statistics of meteorological variables at different time scales.

Multi-Agent simulation of generation capacity expansion decisions.

DOE Office of Scientific and Technical Information (OSTI.GOV)

Botterud, A.; Mahalik, M.; Conzelmann, G.

2008-01-01

In this paper, we use a multi-agent simulation model, EMCAS, to analyze generation expansion in the Iberian electricity market. The expansion model simulates generation investment decisions of decentralized generating companies (GenCos) interacting in a complex, multidimensional environment. A probabilistic dispatch algorithm calculates prices and profits for new candidate units in different future states of the system. Uncertainties in future load, hydropower conditions, and competitorspsila actions are represented in a scenario tree, and decision analysis is used to identify the optimal expansion decision for each individual GenCo. We run the model using detailed data for the Iberian market. In a scenariomore » analysis, we look at the impact of market design variables, such as the energy price cap and carbon emission prices. We also analyze how market concentration and GenCospsila risk preferences influence the timing and choice of new generating capacity.« less

Global/local methods for probabilistic structural analysis

NASA Technical Reports Server (NTRS)

Millwater, H. R.; Wu, Y.-T.

1993-01-01

A probabilistic global/local method is proposed to reduce the computational requirements of probabilistic structural analysis. A coarser global model is used for most of the computations with a local more refined model used only at key probabilistic conditions. The global model is used to establish the cumulative distribution function (cdf) and the Most Probable Point (MPP). The local model then uses the predicted MPP to adjust the cdf value. The global/local method is used within the advanced mean value probabilistic algorithm. The local model can be more refined with respect to the g1obal model in terms of finer mesh, smaller time step, tighter tolerances, etc. and can be used with linear or nonlinear models. The basis for this approach is described in terms of the correlation between the global and local models which can be estimated from the global and local MPPs. A numerical example is presented using the NESSUS probabilistic structural analysis program with the finite element method used for the structural modeling. The results clearly indicate a significant computer savings with minimal loss in accuracy.

Global/local methods for probabilistic structural analysis

NASA Astrophysics Data System (ADS)

Millwater, H. R.; Wu, Y.-T.

1993-04-01

A probabilistic global/local method is proposed to reduce the computational requirements of probabilistic structural analysis. A coarser global model is used for most of the computations with a local more refined model used only at key probabilistic conditions. The global model is used to establish the cumulative distribution function (cdf) and the Most Probable Point (MPP). The local model then uses the predicted MPP to adjust the cdf value. The global/local method is used within the advanced mean value probabilistic algorithm. The local model can be more refined with respect to the g1obal model in terms of finer mesh, smaller time step, tighter tolerances, etc. and can be used with linear or nonlinear models. The basis for this approach is described in terms of the correlation between the global and local models which can be estimated from the global and local MPPs. A numerical example is presented using the NESSUS probabilistic structural analysis program with the finite element method used for the structural modeling. The results clearly indicate a significant computer savings with minimal loss in accuracy.

A probabilistic topic model for clinical risk stratification from electronic health records.

PubMed

Huang, Zhengxing; Dong, Wei; Duan, Huilong

2015-12-01

Risk stratification aims to provide physicians with the accurate assessment of a patient's clinical risk such that an individualized prevention or management strategy can be developed and delivered. Existing risk stratification techniques mainly focus on predicting the overall risk of an individual patient in a supervised manner, and, at the cohort level, often offer little insight beyond a flat score-based segmentation from the labeled clinical dataset. To this end, in this paper, we propose a new approach for risk stratification by exploring a large volume of electronic health records (EHRs) in an unsupervised fashion. Along this line, this paper proposes a novel probabilistic topic modeling framework called probabilistic risk stratification model (PRSM) based on Latent Dirichlet Allocation (LDA). The proposed PRSM recognizes a patient clinical state as a probabilistic combination of latent sub-profiles, and generates sub-profile-specific risk tiers of patients from their EHRs in a fully unsupervised fashion. The achieved stratification results can be easily recognized as high-, medium- and low-risk, respectively. In addition, we present an extension of PRSM, called weakly supervised PRSM (WS-PRSM) by incorporating minimum prior information into the model, in order to improve the risk stratification accuracy, and to make our models highly portable to risk stratification tasks of various diseases. We verify the effectiveness of the proposed approach on a clinical dataset containing 3463 coronary heart disease (CHD) patient instances. Both PRSM and WS-PRSM were compared with two established supervised risk stratification algorithms, i.e., logistic regression and support vector machine, and showed the effectiveness of our models in risk stratification of CHD in terms of the Area Under the receiver operating characteristic Curve (AUC) analysis. As well, in comparison with PRSM, WS-PRSM has over 2% performance gain, on the experimental dataset, demonstrating that incorporating risk scoring knowledge as prior information can improve the performance in risk stratification. Experimental results reveal that our models achieve competitive performance in risk stratification in comparison with existing supervised approaches. In addition, the unsupervised nature of our models makes them highly portable to the risk stratification tasks of various diseases. Moreover, patient sub-profiles and sub-profile-specific risk tiers generated by our models are coherent and informative, and provide significant potential to be explored for the further tasks, such as patient cohort analysis. We hypothesize that the proposed framework can readily meet the demand for risk stratification from a large volume of EHRs in an open-ended fashion. Copyright © 2015 Elsevier Inc. All rights reserved.

Improving medium-range ensemble streamflow forecasts through statistical post-processing

NASA Astrophysics Data System (ADS)

Mendoza, Pablo; Wood, Andy; Clark, Elizabeth; Nijssen, Bart; Clark, Martyn; Ramos, Maria-Helena; Nowak, Kenneth; Arnold, Jeffrey

2017-04-01

Probabilistic hydrologic forecasts are a powerful source of information for decision-making in water resources operations. A common approach is the hydrologic model-based generation of streamflow forecast ensembles, which can be implemented to account for different sources of uncertainties - e.g., from initial hydrologic conditions (IHCs), weather forecasts, and hydrologic model structure and parameters. In practice, hydrologic ensemble forecasts typically have biases and spread errors stemming from errors in the aforementioned elements, resulting in a degradation of probabilistic properties. In this work, we compare several statistical post-processing techniques applied to medium-range ensemble streamflow forecasts obtained with the System for Hydromet Applications, Research and Prediction (SHARP). SHARP is a fully automated prediction system for the assessment and demonstration of short-term to seasonal streamflow forecasting applications, developed by the National Center for Atmospheric Research, University of Washington, U.S. Army Corps of Engineers, and U.S. Bureau of Reclamation. The suite of post-processing techniques includes linear blending, quantile mapping, extended logistic regression, quantile regression, ensemble analogs, and the generalized linear model post-processor (GLMPP). We assess and compare these techniques using multi-year hindcasts in several river basins in the western US. This presentation discusses preliminary findings about the effectiveness of the techniques for improving probabilistic skill, reliability, discrimination, sharpness and resolution.

Probabilistic Analysis of a Composite Crew Module

NASA Technical Reports Server (NTRS)

Mason, Brian H.; Krishnamurthy, Thiagarajan

2011-01-01

An approach for conducting reliability-based analysis (RBA) of a Composite Crew Module (CCM) is presented. The goal is to identify and quantify the benefits of probabilistic design methods for the CCM and future space vehicles. The coarse finite element model from a previous NASA Engineering and Safety Center (NESC) project is used as the baseline deterministic analysis model to evaluate the performance of the CCM using a strength-based failure index. The first step in the probabilistic analysis process is the determination of the uncertainty distributions for key parameters in the model. Analytical data from water landing simulations are used to develop an uncertainty distribution, but such data were unavailable for other load cases. The uncertainty distributions for the other load scale factors and the strength allowables are generated based on assumed coefficients of variation. Probability of first-ply failure is estimated using three methods: the first order reliability method (FORM), Monte Carlo simulation, and conditional sampling. Results for the three methods were consistent. The reliability is shown to be driven by first ply failure in one region of the CCM at the high altitude abort load set. The final predicted probability of failure is on the order of 10-11 due to the conservative nature of the factors of safety on the deterministic loads.

Probabilistic modeling approach for evaluating the compliance of ready-to-eat foods with new European Union safety criteria for Listeria monocytogenes.

PubMed

Koutsoumanis, Konstantinos; Angelidis, Apostolos S

2007-08-01

Among the new microbiological criteria that have been incorporated in EU Regulation 2073/2005, of particular interest are those concerning Listeria monocytogenes in ready-to eat (RTE) foods, because for certain food categories, they no longer require zero tolerance but rather specify a maximum allowable concentration of 100 CFU/g or ml. This study presents a probabilistic modeling approach for evaluating the compliance of RTE sliced meat products with the new safety criteria for L. monocytogenes. The approach was based on the combined use of (i) growth/no growth boundary models, (ii) kinetic growth models, (iii) product characteristics data (pH, a(w), shelf life) collected from 160 meat products from the Hellenic retail market, and (iv) storage temperature data recorded from 50 retail stores in Greece. This study shows that probabilistic analysis of the above components using Monte Carlo simulation, which takes into account the variability of factors affecting microbial growth, can lead to a realistic estimation of the behavior of L. monocytogenes throughout the food supply chain, and the quantitative output generated can be further used by food managers as a decision-making tool regarding the design or modification of a product's formulation or its "use-by" date in order to ensure its compliance with the new safety criteria. The study also argues that compliance of RTE foods with the new safety criteria should not be considered a parameter with a discrete and binary outcome because it depends on factors such as product characteristics, storage temperature, and initial contamination level, which display considerable variability even among different packages of the same RTE product. Rather, compliance should be expressed and therefore regulated in a more probabilistic fashion.

Comparison of numerical weather prediction based deterministic and probabilistic wind resource assessment methods

DOE Office of Scientific and Technical Information (OSTI.GOV)

Zhang, Jie; Draxl, Caroline; Hopson, Thomas

Numerical weather prediction (NWP) models have been widely used for wind resource assessment. Model runs with higher spatial resolution are generally more accurate, yet extremely computational expensive. An alternative approach is to use data generated by a low resolution NWP model, in conjunction with statistical methods. In order to analyze the accuracy and computational efficiency of different types of NWP-based wind resource assessment methods, this paper performs a comparison of three deterministic and probabilistic NWP-based wind resource assessment methodologies: (i) a coarse resolution (0.5 degrees x 0.67 degrees) global reanalysis data set, the Modern-Era Retrospective Analysis for Research and Applicationsmore » (MERRA); (ii) an analog ensemble methodology based on the MERRA, which provides both deterministic and probabilistic predictions; and (iii) a fine resolution (2-km) NWP data set, the Wind Integration National Dataset (WIND) Toolkit, based on the Weather Research and Forecasting model. Results show that: (i) as expected, the analog ensemble and WIND Toolkit perform significantly better than MERRA confirming their ability to downscale coarse estimates; (ii) the analog ensemble provides the best estimate of the multi-year wind distribution at seven of the nine sites, while the WIND Toolkit is the best at one site; (iii) the WIND Toolkit is more accurate in estimating the distribution of hourly wind speed differences, which characterizes the wind variability, at five of the available sites, with the analog ensemble being best at the remaining four locations; and (iv) the analog ensemble computational cost is negligible, whereas the WIND Toolkit requires large computational resources. Future efforts could focus on the combination of the analog ensemble with intermediate resolution (e.g., 10-15 km) NWP estimates, to considerably reduce the computational burden, while providing accurate deterministic estimates and reliable probabilistic assessments.« less

Computation of probabilistic hazard maps and source parameter estimation for volcanic ash transport and dispersion

DOE Office of Scientific and Technical Information (OSTI.GOV)

Madankan, R.; Pouget, S.; Singla, P., E-mail: psingla@buffalo.edu

Volcanic ash advisory centers are charged with forecasting the movement of volcanic ash plumes, for aviation, health and safety preparation. Deterministic mathematical equations model the advection and dispersion of these plumes. However initial plume conditions – height, profile of particle location, volcanic vent parameters – are known only approximately at best, and other features of the governing system such as the windfield are stochastic. These uncertainties make forecasting plume motion difficult. As a result of these uncertainties, ash advisories based on a deterministic approach tend to be conservative, and many times over/under estimate the extent of a plume. This papermore » presents an end-to-end framework for generating a probabilistic approach to ash plume forecasting. This framework uses an ensemble of solutions, guided by Conjugate Unscented Transform (CUT) method for evaluating expectation integrals. This ensemble is used to construct a polynomial chaos expansion that can be sampled cheaply, to provide a probabilistic model forecast. The CUT method is then combined with a minimum variance condition, to provide a full posterior pdf of the uncertain source parameters, based on observed satellite imagery. The April 2010 eruption of the Eyjafjallajökull volcano in Iceland is employed as a test example. The puff advection/dispersion model is used to hindcast the motion of the ash plume through time, concentrating on the period 14–16 April 2010. Variability in the height and particle loading of that eruption is introduced through a volcano column model called bent. Output uncertainty due to the assumed uncertain input parameter probability distributions, and a probabilistic spatial-temporal estimate of ash presence are computed.« less

Development of a coupled hydrological - hydrodynamic model for probabilistic catchment flood inundation modelling

NASA Astrophysics Data System (ADS)

Quinn, Niall; Freer, Jim; Coxon, Gemma; Dunne, Toby; Neal, Jeff; Bates, Paul; Sampson, Chris; Smith, Andy; Parkin, Geoff

2017-04-01

Computationally efficient flood inundation modelling systems capable of representing important hydrological and hydrodynamic flood generating processes over relatively large regions are vital for those interested in flood preparation, response, and real time forecasting. However, such systems are currently not readily available. This can be particularly important where flood predictions from intense rainfall are considered as the processes leading to flooding often involve localised, non-linear spatially connected hillslope-catchment responses. Therefore, this research introduces a novel hydrological-hydraulic modelling framework for the provision of probabilistic flood inundation predictions across catchment to regional scales that explicitly account for spatial variability in rainfall-runoff and routing processes. Approaches have been developed to automate the provision of required input datasets and estimate essential catchment characteristics from freely available, national datasets. This is an essential component of the framework as when making predictions over multiple catchments or at relatively large scales, and where data is often scarce, obtaining local information and manually incorporating it into the model quickly becomes infeasible. An extreme flooding event in the town of Morpeth, NE England, in 2008 was used as a first case study evaluation of the modelling framework introduced. The results demonstrated a high degree of prediction accuracy when comparing modelled and reconstructed event characteristics for the event, while the efficiency of the modelling approach used enabled the generation of relatively large ensembles of realisations from which uncertainty within the prediction may be represented. This research supports previous literature highlighting the importance of probabilistic forecasting, particularly during extreme events, which can be often be poorly characterised or even missed by deterministic predictions due to the inherent uncertainty in any model application. Future research will aim to further evaluate the robustness of the approaches introduced by applying the modelling framework to a variety of historical flood events across UK catchments. Furthermore, the flexibility and efficiency of the framework is ideally suited to the examination of the propagation of errors through the model which will help gain a better understanding of the dominant sources of uncertainty currently impacting flood inundation predictions.

BA3b and BA1 activate in a serial fashion after median nerve stimulation: direct evidence from combining source analysis of evoked fields and cytoarchitectonic probabilistic maps.

PubMed

Papadelis, Christos; Eickhoff, Simon B; Zilles, Karl; Ioannides, Andreas A

2011-01-01

This study combines source analysis imaging data for early somatosensory processing and the probabilistic cytoarchitectonic maps (PCMs). Human somatosensory evoked fields (SEFs) were recorded by stimulating left and right median nerves. Filtering the recorded responses in different frequency ranges identified the most responsive frequency band. The short-latency averaged SEFs were analyzed using a single equivalent current dipole (ECD) model and magnetic field tomography (MFT). The identified foci of activity were superimposed with PCMs. Two major components of opposite polarity were prominent around 21 and 31 ms. A weak component around 25 ms was also identified. For the most responsive frequency band (50-150 Hz) ECD and MFT revealed one focal source at the contralateral Brodmann area 3b (BA3b) at the peak of N20. The component ~25 ms was localised in Brodmann area 1 (BA1) in 50-150 Hz. By using ECD, focal generators around 28-30 ms located initially in BA3b and 2 ms later to BA1. MFT also revealed two focal sources - one in BA3b and one in BA1 for these latencies. Our results provide direct evidence that the earliest cortical response after median nerve stimulation is generated within the contralateral BA3b. BA1 activation few milliseconds later indicates a serial mode of somatosensory processing within cytoarchitectonic SI subdivisions. Analysis of non-invasive magnetoencephalography (MEG) data and the use of PCMs allow unambiguous and quantitative (probabilistic) interpretation of cytoarchitectonic identity of activated areas following median nerve stimulation, even with the simple ECD model, but only when the model fits the data extremely well. Copyright © 2010 Elsevier Inc. All rights reserved.

Information Measures for Multisensor Systems

DTIC Science & Technology

2013-12-11

permuted to generate spectra that were non- physical but preserved the entropy of the source spectra. Another 1000 spectra were constructed to mimic co...Research Laboratory (NRL) has yielded probabilistic models for spectral data that enable the computation of information measures such as entropy and...22308 Chemical sensing Information theory Spectral data Information entropy Information divergence Mass spectrometry Infrared spectroscopy Multisensor

Development of optimization-based probabilistic earthquake scenarios for the city of Tehran

NASA Astrophysics Data System (ADS)

Zolfaghari, M. R.; Peyghaleh, E.

2016-01-01

This paper presents the methodology and practical example for the application of optimization process to select earthquake scenarios which best represent probabilistic earthquake hazard in a given region. The method is based on simulation of a large dataset of potential earthquakes, representing the long-term seismotectonic characteristics in a given region. The simulation process uses Monte-Carlo simulation and regional seismogenic source parameters to generate a synthetic earthquake catalogue consisting of a large number of earthquakes, each characterized with magnitude, location, focal depth and fault characteristics. Such catalogue provides full distributions of events in time, space and size; however, demands large computation power when is used for risk assessment, particularly when other sources of uncertainties are involved in the process. To reduce the number of selected earthquake scenarios, a mixed-integer linear program formulation is developed in this study. This approach results in reduced set of optimization-based probabilistic earthquake scenario, while maintaining shape of hazard curves and full probabilistic picture by minimizing the error between hazard curves driven by full and reduced sets of synthetic earthquake scenarios. To test the model, the regional seismotectonic and seismogenic characteristics of northern Iran are used to simulate a set of 10,000-year worth of events consisting of some 84,000 earthquakes. The optimization model is then performed multiple times with various input data, taking into account probabilistic seismic hazard for Tehran city as the main constrains. The sensitivity of the selected scenarios to the user-specified site/return period error-weight is also assessed. The methodology could enhance run time process for full probabilistic earthquake studies like seismic hazard and risk assessment. The reduced set is the representative of the contributions of all possible earthquakes; however, it requires far less computation power. The authors have used this approach for risk assessment towards identification of effectiveness-profitability of risk mitigation measures, using optimization model for resource allocation. Based on the error-computation trade-off, 62-earthquake scenarios are chosen to be used for this purpose.

A comparison of metrics for assessing state-of-the-art climate models and implications for probabilistic projections of climate change

NASA Astrophysics Data System (ADS)

Ring, Christoph; Pollinger, Felix; Kaspar-Ott, Irena; Hertig, Elke; Jacobeit, Jucundus; Paeth, Heiko

2018-03-01

A major task of climate science are reliable projections of climate change for the future. To enable more solid statements and to decrease the range of uncertainty, global general circulation models and regional climate models are evaluated based on a 2 × 2 contingency table approach to generate model weights. These weights are compared among different methodologies and their impact on probabilistic projections of temperature and precipitation changes is investigated. Simulated seasonal precipitation and temperature for both 50-year trends and climatological means are assessed at two spatial scales: in seven study regions around the globe and in eight sub-regions of the Mediterranean area. Overall, 24 models of phase 3 and 38 models of phase 5 of the Coupled Model Intercomparison Project altogether 159 transient simulations of precipitation and 119 of temperature from four emissions scenarios are evaluated against the ERA-20C reanalysis over the 20th century. The results show high conformity with previous model evaluation studies. The metrics reveal that mean of precipitation and both temperature mean and trend agree well with the reference dataset and indicate improvement for the more recent ensemble mean, especially for temperature. The method is highly transferrable to a variety of further applications in climate science. Overall, there are regional differences of simulation quality, however, these are less pronounced than those between the results for 50-year mean and trend. The trend results are suitable for assigning weighting factors to climate models. Yet, the implications for probabilistic climate projections is strictly dependent on the region and season.

Climatological Observations for Maritime Prediction and Analysis Support Service (COMPASS)

NASA Astrophysics Data System (ADS)

OConnor, A.; Kirtman, B. P.; Harrison, S.; Gorman, J.

2016-02-01

Current US Navy forecasting systems cannot easily incorporate extended-range forecasts that can improve mission readiness and effectiveness; ensure safety; and reduce cost, labor, and resource requirements. If Navy operational planners had systems that incorporated these forecasts, they could plan missions using more reliable and longer-term weather and climate predictions. Further, using multi-model forecast ensembles instead of single forecasts would produce higher predictive performance. Extended-range multi-model forecast ensembles, such as those available in the North American Multi-Model Ensemble (NMME), are ideal for system integration because of their high skill predictions; however, even higher skill predictions can be produced if forecast model ensembles are combined correctly. While many methods for weighting models exist, the best method in a given environment requires expert knowledge of the models and combination methods.We present an innovative approach that uses machine learning to combine extended-range predictions from multi-model forecast ensembles and generate a probabilistic forecast for any region of the globe up to 12 months in advance. Our machine-learning approach uses 30 years of hindcast predictions to learn patterns of forecast model successes and failures. Each model is assigned a weight for each environmental condition, 100 km2 region, and day given any expected environmental information. These weights are then applied to the respective predictions for the region and time of interest to effectively stitch together a single, coherent probabilistic forecast. Our experimental results demonstrate the benefits of our approach to produce extended-range probabilistic forecasts for regions and time periods of interest that are superior, in terms of skill, to individual NMME forecast models and commonly weighted models. The probabilistic forecast leverages the strengths of three NMME forecast models to predict environmental conditions for an area spanning from San Diego, CA to Honolulu, HI, seven months in-advance. Key findings include: weighted combinations of models are strictly better than individual models; machine-learned combinations are especially better; and forecasts produced using our approach have the highest rank probability skill score most often.

A neural network model of metaphor understanding with dynamic interaction based on a statistical language analysis: targeting a human-like model.

PubMed

Terai, Asuka; Nakagawa, Masanori

2007-08-01

The purpose of this paper is to construct a model that represents the human process of understanding metaphors, focusing specifically on similes of the form an "A like B". Generally speaking, human beings are able to generate and understand many sorts of metaphors. This study constructs the model based on a probabilistic knowledge structure for concepts which is computed from a statistical analysis of a large-scale corpus. Consequently, this model is able to cover the many kinds of metaphors that human beings can generate. Moreover, the model implements the dynamic process of metaphor understanding by using a neural network with dynamic interactions. Finally, the validity of the model is confirmed by comparing model simulations with the results from a psychological experiment.

A spatio-temporal model for probabilistic seismic hazard zonation of Tehran

NASA Astrophysics Data System (ADS)

Hashemi, Mahdi; Alesheikh, Ali Asghar; Zolfaghari, Mohammad Reza

2013-08-01

A precondition for all disaster management steps, building damage prediction, and construction code developments is a hazard assessment that shows the exceedance probabilities of different ground motion levels at a site considering different near- and far-field earthquake sources. The seismic sources are usually categorized as time-independent area sources and time-dependent fault sources. While the earlier incorporates the small and medium events, the later takes into account only the large characteristic earthquakes. In this article, a probabilistic approach is proposed to aggregate the effects of time-dependent and time-independent sources on seismic hazard. The methodology is then applied to generate three probabilistic seismic hazard maps of Tehran for 10%, 5%, and 2% exceedance probabilities in 50 years. The results indicate an increase in peak ground acceleration (PGA) values toward the southeastern part of the study area and the PGA variations are mostly controlled by the shear wave velocities across the city. In addition, the implementation of the methodology takes advantage of GIS capabilities especially raster-based analyses and representations. During the estimation of the PGA exceedance rates, the emphasis has been placed on incorporating the effects of different attenuation relationships and seismic source models by using a logic tree.

Learning In networks

NASA Technical Reports Server (NTRS)

Buntine, Wray L.

1995-01-01

Intelligent systems require software incorporating probabilistic reasoning, and often times learning. Networks provide a framework and methodology for creating this kind of software. This paper introduces network models based on chain graphs with deterministic nodes. Chain graphs are defined as a hierarchical combination of Bayesian and Markov networks. To model learning, plates on chain graphs are introduced to model independent samples. The paper concludes by discussing various operations that can be performed on chain graphs with plates as a simplification process or to generate learning algorithms.

A probabilistic framework for single-sensor acoustic emission source localization in thin metallic plates

NASA Astrophysics Data System (ADS)

Ebrahimkhanlou, Arvin; Salamone, Salvatore

2017-09-01

Tracking edge-reflected acoustic emission (AE) waves can allow the localization of their sources. Specifically, in bounded isotropic plate structures, only one sensor may be used to perform these source localizations. The primary goal of this paper is to develop a three-step probabilistic framework to quantify the uncertainties associated with such single-sensor localizations. According to this framework, a probabilistic approach is first used to estimate the direct distances between AE sources and the sensor. Then, an analytical model is used to reconstruct the envelope of edge-reflected AE signals based on the source-to-sensor distance estimations and their first arrivals. Finally, the correlation between the probabilistically reconstructed envelopes and recorded AE signals are used to estimate confidence contours for the location of AE sources. To validate the proposed framework, Hsu-Nielsen pencil lead break (PLB) tests were performed on the surface as well as the edges of an aluminum plate. The localization results show that the estimated confidence contours surround the actual source locations. In addition, the performance of the framework was tested in a noisy environment simulated by two dummy transducers and an arbitrary wave generator. The results show that in low-noise environments, the shape and size of the confidence contours depend on the sources and their locations. However, at highly noisy environments, the size of the confidence contours monotonically increases with the noise floor. Such probabilistic results suggest that the proposed probabilistic framework could thus provide more comprehensive information regarding the location of AE sources.

Development of a Probabilistic Dynamic Synthesis Method for the Analysis of Nondeterministic Structures

NASA Technical Reports Server (NTRS)

Brown, A. M.

1998-01-01

Accounting for the statistical geometric and material variability of structures in analysis has been a topic of considerable research for the last 30 years. The determination of quantifiable measures of statistical probability of a desired response variable, such as natural frequency, maximum displacement, or stress, to replace experience-based "safety factors" has been a primary goal of these studies. There are, however, several problems associated with their satisfactory application to realistic structures, such as bladed disks in turbomachinery. These include the accurate definition of the input random variables (rv's), the large size of the finite element models frequently used to simulate these structures, which makes even a single deterministic analysis expensive, and accurate generation of the cumulative distribution function (CDF) necessary to obtain the probability of the desired response variables. The research presented here applies a methodology called probabilistic dynamic synthesis (PDS) to solve these problems. The PDS method uses dynamic characteristics of substructures measured from modal test as the input rv's, rather than "primitive" rv's such as material or geometric uncertainties. These dynamic characteristics, which are the free-free eigenvalues, eigenvectors, and residual flexibility (RF), are readily measured and for many substructures, a reasonable sample set of these measurements can be obtained. The statistics for these rv's accurately account for the entire random character of the substructure. Using the RF method of component mode synthesis, these dynamic characteristics are used to generate reduced-size sample models of the substructures, which are then coupled to form system models. These sample models are used to obtain the CDF of the response variable by either applying Monte Carlo simulation or by generating data points for use in the response surface reliability method, which can perform the probabilistic analysis with an order of magnitude less computational effort. Both free- and forced-response analyses have been performed, and the results indicate that, while there is considerable room for improvement, the method produces usable and more representative solutions for the design of realistic structures with a substantial savings in computer time.

Probabilistic Radiological Performance Assessment Modeling and Uncertainty

NASA Astrophysics Data System (ADS)

Tauxe, J.

2004-12-01

A generic probabilistic radiological Performance Assessment (PA) model is presented. The model, built using the GoldSim systems simulation software platform, concerns contaminant transport and dose estimation in support of decision making with uncertainty. Both the U.S. Nuclear Regulatory Commission (NRC) and the U.S. Department of Energy (DOE) require assessments of potential future risk to human receptors of disposal of LLW. Commercially operated LLW disposal facilities are licensed by the NRC (or agreement states), and the DOE operates such facilities for disposal of DOE-generated LLW. The type of PA model presented is probabilistic in nature, and hence reflects the current state of knowledge about the site by using probability distributions to capture what is expected (central tendency or average) and the uncertainty (e.g., standard deviation) associated with input parameters, and propagating through the model to arrive at output distributions that reflect expected performance and the overall uncertainty in the system. Estimates of contaminant release rates, concentrations in environmental media, and resulting doses to human receptors well into the future are made by running the model in Monte Carlo fashion, with each realization representing a possible combination of input parameter values. Statistical summaries of the results can be compared to regulatory performance objectives, and decision makers are better informed of the inherently uncertain aspects of the model which supports their decision-making. While this information may make some regulators uncomfortable, they must realize that uncertainties which were hidden in a deterministic analysis are revealed in a probabilistic analysis, and the chance of making a correct decision is now known rather than hoped for. The model includes many typical features and processes that would be part of a PA, but is entirely fictitious. This does not represent any particular site and is meant to be a generic example. A practitioner could, however, start with this model as a GoldSim template and, by adding site specific features and parameter values (distributions), use this model as a starting point for a real model to be used in real decision making.

Stochastic Controls on Nitrate Transport and Cycling

NASA Astrophysics Data System (ADS)

Botter, G.; Settin, T.; Alessi Celegon, E.; Marani, M.; Rinaldo, A.

2005-12-01

In this paper, the impact of nutrient inputs on basin-scale nitrates losses is investigated in a probabilistic framework by means of a continuous, geomorphologically based, Montecarlo approach, which explicitly tackles the random character of the processes controlling nitrates generation, transformation and transport in river basins. This is obtained by coupling the stochastic generation of climatic and rainfall series with simplified hydrologic and biogeochemical models operating at the hillslope scale. Special attention is devoted to the spatial and temporal variability of nitrogen sources of agricultural origin and to the effect of temporally distributed rainfall fields on the ensuing nitrates leaching. The influence of random climatic variables on bio-geochemical processes affecting the nitrogen cycle in the soil-water system (e.g. plant uptake, nitrification and denitrification, mineralization), is also considered. The approach developed has been applied to a catchment located in North-Eastern Italy and is used to provide probabilistic estimates of the NO_3 load transferred downstream, which is received and accumulated in the Venice lagoon. We found that the nitrogen load introduced by fertilizations significantly affects the pdf of the nitrates content in the soil moisture, leading to prolonged risks of increased nitrates leaching from soil. The model allowed the estimation of the impact of different practices on the probabilistic structure of the basin-scale hydrologic and chemical response. As a result, the return period of the water volumes and of the nitrates loads released into the Venice lagoon has been linked directly to the ongoing climatic, pluviometric and agricultural regimes, with relevant implications for environmental planning activities aimed at achieving sustainable management practices.

Stochastic and Statistical Analysis of Utility Revenues and Weather Data Analysis for Consumer Demand Estimation in Smart Grids

PubMed Central

Ali, S. M.; Mehmood, C. A; Khan, B.; Jawad, M.; Farid, U; Jadoon, J. K.; Ali, M.; Tareen, N. K.; Usman, S.; Majid, M.; Anwar, S. M.

2016-01-01

In smart grid paradigm, the consumer demands are random and time-dependent, owning towards stochastic probabilities. The stochastically varying consumer demands have put the policy makers and supplying agencies in a demanding position for optimal generation management. The utility revenue functions are highly dependent on the consumer deterministic stochastic demand models. The sudden drifts in weather parameters effects the living standards of the consumers that in turn influence the power demands. Considering above, we analyzed stochastically and statistically the effect of random consumer demands on the fixed and variable revenues of the electrical utilities. Our work presented the Multi-Variate Gaussian Distribution Function (MVGDF) probabilistic model of the utility revenues with time-dependent consumer random demands. Moreover, the Gaussian probabilities outcome of the utility revenues is based on the varying consumer n demands data-pattern. Furthermore, Standard Monte Carlo (SMC) simulations are performed that validated the factor of accuracy in the aforesaid probabilistic demand-revenue model. We critically analyzed the effect of weather data parameters on consumer demands using correlation and multi-linear regression schemes. The statistical analysis of consumer demands provided a relationship between dependent (demand) and independent variables (weather data) for utility load management, generation control, and network expansion. PMID:27314229

Stochastic and Statistical Analysis of Utility Revenues and Weather Data Analysis for Consumer Demand Estimation in Smart Grids.

PubMed

Ali, S M; Mehmood, C A; Khan, B; Jawad, M; Farid, U; Jadoon, J K; Ali, M; Tareen, N K; Usman, S; Majid, M; Anwar, S M

2016-01-01

In smart grid paradigm, the consumer demands are random and time-dependent, owning towards stochastic probabilities. The stochastically varying consumer demands have put the policy makers and supplying agencies in a demanding position for optimal generation management. The utility revenue functions are highly dependent on the consumer deterministic stochastic demand models. The sudden drifts in weather parameters effects the living standards of the consumers that in turn influence the power demands. Considering above, we analyzed stochastically and statistically the effect of random consumer demands on the fixed and variable revenues of the electrical utilities. Our work presented the Multi-Variate Gaussian Distribution Function (MVGDF) probabilistic model of the utility revenues with time-dependent consumer random demands. Moreover, the Gaussian probabilities outcome of the utility revenues is based on the varying consumer n demands data-pattern. Furthermore, Standard Monte Carlo (SMC) simulations are performed that validated the factor of accuracy in the aforesaid probabilistic demand-revenue model. We critically analyzed the effect of weather data parameters on consumer demands using correlation and multi-linear regression schemes. The statistical analysis of consumer demands provided a relationship between dependent (demand) and independent variables (weather data) for utility load management, generation control, and network expansion.

Comparison of Deterministic and Probabilistic Radial Distribution Systems Load Flow

NASA Astrophysics Data System (ADS)

Gupta, Atma Ram; Kumar, Ashwani

2017-12-01

Distribution system network today is facing the challenge of meeting increased load demands from the industrial, commercial and residential sectors. The pattern of load is highly dependent on consumer behavior and temporal factors such as season of the year, day of the week or time of the day. For deterministic radial distribution load flow studies load is taken as constant. But, load varies continually with a high degree of uncertainty. So, there is a need to model probable realistic load. Monte-Carlo Simulation is used to model the probable realistic load by generating random values of active and reactive power load from the mean and standard deviation of the load and for solving a Deterministic Radial Load Flow with these values. The probabilistic solution is reconstructed from deterministic data obtained for each simulation. The main contribution of the work is: Finding impact of probable realistic ZIP load modeling on balanced radial distribution load flow. Finding impact of probable realistic ZIP load modeling on unbalanced radial distribution load flow. Compare the voltage profile and losses with probable realistic ZIP load modeling for balanced and unbalanced radial distribution load flow.

Soft Mixer Assignment in a Hierarchical Generative Model of Natural Scene Statistics

PubMed Central

Schwartz, Odelia; Sejnowski, Terrence J.; Dayan, Peter

2010-01-01

Gaussian scale mixture models offer a top-down description of signal generation that captures key bottom-up statistical characteristics of filter responses to images. However, the pattern of dependence among the filters for this class of models is prespecified. We propose a novel extension to the gaussian scale mixture model that learns the pattern of dependence from observed inputs and thereby induces a hierarchical representation of these inputs. Specifically, we propose that inputs are generated by gaussian variables (modeling local filter structure), multiplied by a mixer variable that is assigned probabilistically to each input from a set of possible mixers. We demonstrate inference of both components of the generative model, for synthesized data and for different classes of natural images, such as a generic ensemble and faces. For natural images, the mixer variable assignments show invariances resembling those of complex cells in visual cortex; the statistics of the gaussian components of the model are in accord with the outputs of divisive normalization models. We also show how our model helps interrelate a wide range of models of image statistics and cortical processing. PMID:16999575

Extended Range Prediction of Indian Summer Monsoon: Current status

NASA Astrophysics Data System (ADS)

Sahai, A. K.; Abhilash, S.; Borah, N.; Joseph, S.; Chattopadhyay, R.; S, S.; Rajeevan, M.; Mandal, R.; Dey, A.

2014-12-01

The main focus of this study is to develop forecast consensus in the extended range prediction (ERP) of monsoon Intraseasonal oscillations using a suit of different variants of Climate Forecast system (CFS) model. In this CFS based Grand MME prediction system (CGMME), the ensemble members are generated by perturbing the initial condition and using different configurations of CFSv2. This is to address the role of different physical mechanisms known to have control on the error growth in the ERP in the 15-20 day time scale. The final formulation of CGMME is based on 21 ensembles of the standalone Global Forecast System (GFS) forced with bias corrected forecasted SST from CFS, 11 low resolution CFST126 and 11 high resolution CFST382. Thus, we develop the multi-model consensus forecast for the ERP of Indian summer monsoon (ISM) using a suite of different variants of CFS model. This coordinated international effort lead towards the development of specific tailor made regional forecast products over Indian region. Skill of deterministic and probabilistic categorical rainfall forecast as well the verification of large-scale low frequency monsoon intraseasonal oscillations has been carried out using hindcast from 2001-2012 during the monsoon season in which all models are initialized at every five days starting from 16May to 28 September. The skill of deterministic forecast from CGMME is better than the best participating single model ensemble configuration (SME). The CGMME approach is believed to quantify the uncertainty in both initial conditions and model formulation. Main improvement is attained in probabilistic forecast which is because of an increase in the ensemble spread, thereby reducing the error due to over-confident ensembles in a single model configuration. For probabilistic forecast, three tercile ranges are determined by ranking method based on the percentage of ensemble members from all the participating models falls in those three categories. CGMME further added value to both deterministic and probability forecast compared to raw SME's and this better skill is probably flows from large spread and improved spread-error relationship. CGMME system is currently capable of generating ER prediction in real time and successfully delivering its experimental operational ER forecast of ISM for the last few years.

Visualizing Uncertainty for Probabilistic Weather Forecasting based on Reforecast Analogs

NASA Astrophysics Data System (ADS)

Pelorosso, Leandro; Diehl, Alexandra; Matković, Krešimir; Delrieux, Claudio; Ruiz, Juan; Gröeller, M. Eduard; Bruckner, Stefan

2016-04-01

Numerical weather forecasts are prone to uncertainty coming from inaccuracies in the initial and boundary conditions and lack of precision in numerical models. Ensemble of forecasts partially addresses these problems by considering several runs of the numerical model. Each forecast is generated with different initial and boundary conditions and different model configurations [GR05]. The ensembles can be expressed as probabilistic forecasts, which have proven to be very effective in the decision-making processes [DE06]. The ensemble of forecasts represents only some of the possible future atmospheric states, usually underestimating the degree of uncertainty in the predictions [KAL03, PH06]. Hamill and Whitaker [HW06] introduced the "Reforecast Analog Regression" (RAR) technique to overcome the limitations of ensemble forecasting. This technique produces probabilistic predictions based on the analysis of historical forecasts and observations. Visual analytics provides tools for processing, visualizing, and exploring data to get new insights and discover hidden information patterns in an interactive exchange between the user and the application [KMS08]. In this work, we introduce Albero, a visual analytics solution for probabilistic weather forecasting based on the RAR technique. Albero targets at least two different type of users: "forecasters", who are meteorologists working in operational weather forecasting and "researchers", who work in the construction of numerical prediction models. Albero is an efficient tool for analyzing precipitation forecasts, allowing forecasters to make and communicate quick decisions. Our solution facilitates the analysis of a set of probabilistic forecasts, associated statistical data, observations and uncertainty. A dashboard with small-multiples of probabilistic forecasts allows the forecasters to analyze at a glance the distribution of probabilities as a function of time, space, and magnitude. It provides the user with a more accurate measure of forecast uncertainty that could result in better decision-making. It offers different level of abstractions to help with the recalibration of the RAR method. It also has an inspection tool that displays the selected analogs, their observations and statistical data. It gives the users access to inner parts of the method, unveiling hidden information. References [GR05] GNEITING T., RAFTERY A. E.: Weather forecasting with ensemble methods. Science 310, 5746, 248-249, 2005. [KAL03] KALNAY E.: Atmospheric modeling, data assimilation and predictability. Cambridge University Press, 2003. [PH06] PALMER T., HAGEDORN R.: Predictability of weather and climate. Cambridge University Press, 2006. [HW06] HAMILL T. M., WHITAKER J. S.: Probabilistic quantitative precipitation forecasts based on reforecast analogs: Theory and application. Monthly Weather Review 134, 11, 3209-3229, 2006. [DE06] DEITRICK S., EDSALL R.: The influence of uncertainty visualization on decision making: An empirical evaluation. Springer, 2006. [KMS08] KEIM D. A., MANSMANN F., SCHNEIDEWIND J., THOMAS J., ZIEGLER H.: Visual analytics: Scope and challenges. Springer, 2008.

Development and application of a probabilistic method for wildfire suppression cost modeling

Treesearch

Matthew P. Thompson; Jessica R. Haas; Mark A. Finney; David E. Calkin; Michael S. Hand; Mark J. Browne; Martin Halek; Karen C. Short; Isaac C. Grenfell

2015-01-01

Wildfire activity and escalating suppression costs continue to threaten the financial health of federal land management agencies. In order to minimize and effectively manage the cost of financial risk, agencies need the ability to quantify that risk. A fundamental aim of this research effort, therefore, is to develop a process for generating risk-based metrics for...

Probabilistic material degradation model for aerospace materials subjected to high temperature, mechanical and thermal fatigue, and creep

NASA Technical Reports Server (NTRS)

Boyce, L.

1992-01-01

A probabilistic general material strength degradation model has been developed for structural components of aerospace propulsion systems subjected to diverse random effects. The model has been implemented in two FORTRAN programs, PROMISS (Probabilistic Material Strength Simulator) and PROMISC (Probabilistic Material Strength Calibrator). PROMISS calculates the random lifetime strength of an aerospace propulsion component due to as many as eighteen diverse random effects. Results are presented in the form of probability density functions and cumulative distribution functions of lifetime strength. PROMISC calibrates the model by calculating the values of empirical material constants.

Anticipatory systems using a probabilistic-possibilistic formalism

DOE Office of Scientific and Technical Information (OSTI.GOV)

Tsoukalas, L.H.

1989-01-01

A methodology for the realization of the Anticipatory Paradigm in the diagnosis and control of complex systems, such as power plants, is developed. The objective is to synthesize engineering systems as analogs of certain biological systems which are capable of modifying their present states on the basis of anticipated future states. These future states are construed to be the output of predictive, numerical, stochastic or symbolic models. The mathematical basis of the implementation is developed on the basis of a formulation coupling probabilistic (random) and possibilistic(fuzzy) data in the form of an Information Granule. Random data are generated from observationsmore » and sensors input from the environment. Fuzzy data consists of eqistemic information, such as criteria or constraints qualifying the environmental inputs. The approach generates mathematical performance measures upon which diagnostic inferences and control functions are based. Anticipated performance is generated using a fuzzified Bayes formula. Triplex arithmetic is used in the numerical estimation of the performance measures. Representation of the system is based upon a goal-tree within the rule-based paradigm from the field of Applied Artificial Intelligence. The ensuing construction incorporates a coupling of Symbolic and Procedural programming methods. As a demonstration of the possibility of constructing such systems, a model-based system of a nuclear reactor is constructed. A numerical model of the reactor as a damped simple harmonic oscillator is used. The neutronic behavior is described by a point kinetics model with temperature feedback. The resulting system is programmed in OPS5 for the symbolic component and in FORTRAN for the procedural part.« less

Software for Data Analysis with Graphical Models

NASA Technical Reports Server (NTRS)

Buntine, Wray L.; Roy, H. Scott

1994-01-01

Probabilistic graphical models are being used widely in artificial intelligence and statistics, for instance, in diagnosis and expert systems, as a framework for representing and reasoning with probabilities and independencies. They come with corresponding algorithms for performing statistical inference. This offers a unifying framework for prototyping and/or generating data analysis algorithms from graphical specifications. This paper illustrates the framework with an example and then presents some basic techniques for the task: problem decomposition and the calculation of exact Bayes factors. Other tools already developed, such as automatic differentiation, Gibbs sampling, and use of the EM algorithm, make this a broad basis for the generation of data analysis software.

Reasoning in Reference Games: Individual- vs. Population-Level Probabilistic Modeling

PubMed Central

Franke, Michael; Degen, Judith

2016-01-01

Recent advances in probabilistic pragmatics have achieved considerable success in modeling speakers’ and listeners’ pragmatic reasoning as probabilistic inference. However, these models are usually applied to population-level data, and so implicitly suggest a homogeneous population without individual differences. Here we investigate potential individual differences in Theory-of-Mind related depth of pragmatic reasoning in so-called reference games that require drawing ad hoc Quantity implicatures of varying complexity. We show by Bayesian model comparison that a model that assumes a heterogenous population is a better predictor of our data, especially for comprehension. We discuss the implications for the treatment of individual differences in probabilistic models of language use. PMID:27149675

Quantum-Assisted Learning of Hardware-Embedded Probabilistic Graphical Models

NASA Astrophysics Data System (ADS)

Benedetti, Marcello; Realpe-Gómez, John; Biswas, Rupak; Perdomo-Ortiz, Alejandro

2017-10-01

Mainstream machine-learning techniques such as deep learning and probabilistic programming rely heavily on sampling from generally intractable probability distributions. There is increasing interest in the potential advantages of using quantum computing technologies as sampling engines to speed up these tasks or to make them more effective. However, some pressing challenges in state-of-the-art quantum annealers have to be overcome before we can assess their actual performance. The sparse connectivity, resulting from the local interaction between quantum bits in physical hardware implementations, is considered the most severe limitation to the quality of constructing powerful generative unsupervised machine-learning models. Here, we use embedding techniques to add redundancy to data sets, allowing us to increase the modeling capacity of quantum annealers. We illustrate our findings by training hardware-embedded graphical models on a binarized data set of handwritten digits and two synthetic data sets in experiments with up to 940 quantum bits. Our model can be trained in quantum hardware without full knowledge of the effective parameters specifying the corresponding quantum Gibbs-like distribution; therefore, this approach avoids the need to infer the effective temperature at each iteration, speeding up learning; it also mitigates the effect of noise in the control parameters, making it robust to deviations from the reference Gibbs distribution. Our approach demonstrates the feasibility of using quantum annealers for implementing generative models, and it provides a suitable framework for benchmarking these quantum technologies on machine-learning-related tasks.

Probabilistic graphs as a conceptual and computational tool in hydrology and water management

NASA Astrophysics Data System (ADS)

Schoups, Gerrit

2014-05-01

Originally developed in the fields of machine learning and artificial intelligence, probabilistic graphs constitute a general framework for modeling complex systems in the presence of uncertainty. The framework consists of three components: 1. Representation of the model as a graph (or network), with nodes depicting random variables in the model (e.g. parameters, states, etc), which are joined together by factors. Factors are local probabilistic or deterministic relations between subsets of variables, which, when multiplied together, yield the joint distribution over all variables. 2. Consistent use of probability theory for quantifying uncertainty, relying on basic rules of probability for assimilating data into the model and expressing unknown variables as a function of observations (via the posterior distribution). 3. Efficient, distributed approximation of the posterior distribution using general-purpose algorithms that exploit model structure encoded in the graph. These attributes make probabilistic graphs potentially useful as a conceptual and computational tool in hydrology and water management (and beyond). Conceptually, they can provide a common framework for existing and new probabilistic modeling approaches (e.g. by drawing inspiration from other fields of application), while computationally they can make probabilistic inference feasible in larger hydrological models. The presentation explores, via examples, some of these benefits.

A Point Rainfall Generator With Internal Storm Structure

NASA Astrophysics Data System (ADS)

Marien, J. L.; Vandewiele, G. L.

1986-04-01

A point rainfall generator is a probabilistic model for the time series of rainfall as observed in one geographical point. The main purpose of such a model is to generate long synthetic sequences of rainfall for simulation studies. The present generator is a continuous time model based on 13.5 years of 10-min point rainfalls observed in Belgium and digitized with a resolution of 0.1 mm. The present generator attempts to model all features of the rainfall time series which are important for flood studies as accurately as possible. The original aspects of the model are on the one hand the way in which storms are defined and on the other hand the theoretical model for the internal storm characteristics. The storm definition has the advantage that the important characteristics of successive storms are fully independent and very precisely modelled, even on time bases as small as 10 min. The model of the internal storm characteristics has a strong theoretical structure. This fact justifies better the extrapolation of this model to severe storms for which the data are very sparse. This can be important when using the model to simulate severe flood events.

Towards an improved ensemble precipitation forecast: A probabilistic post-processing approach

NASA Astrophysics Data System (ADS)

Khajehei, Sepideh; Moradkhani, Hamid

2017-03-01

Recently, ensemble post-processing (EPP) has become a commonly used approach for reducing the uncertainty in forcing data and hence hydrologic simulation. The procedure was introduced to build ensemble precipitation forecasts based on the statistical relationship between observations and forecasts. More specifically, the approach relies on a transfer function that is developed based on a bivariate joint distribution between the observations and the simulations in the historical period. The transfer function is used to post-process the forecast. In this study, we propose a Bayesian EPP approach based on copula functions (COP-EPP) to improve the reliability of the precipitation ensemble forecast. Evaluation of the copula-based method is carried out by comparing the performance of the generated ensemble precipitation with the outputs from an existing procedure, i.e. mixed type meta-Gaussian distribution. Monthly precipitation from Climate Forecast System Reanalysis (CFS) and gridded observation from Parameter-Elevation Relationships on Independent Slopes Model (PRISM) have been employed to generate the post-processed ensemble precipitation. Deterministic and probabilistic verification frameworks are utilized in order to evaluate the outputs from the proposed technique. Distribution of seasonal precipitation for the generated ensemble from the copula-based technique is compared to the observation and raw forecasts for three sub-basins located in the Western United States. Results show that both techniques are successful in producing reliable and unbiased ensemble forecast, however, the COP-EPP demonstrates considerable improvement in the ensemble forecast in both deterministic and probabilistic verification, in particular in characterizing the extreme events in wet seasons.

Perception of Risk and Terrorism-Related Behavior Change: Dual Influences of Probabilistic Reasoning and Reality Testing.

PubMed

Denovan, Andrew; Dagnall, Neil; Drinkwater, Kenneth; Parker, Andrew; Clough, Peter

2017-01-01

The present study assessed the degree to which probabilistic reasoning performance and thinking style influenced perception of risk and self-reported levels of terrorism-related behavior change. A sample of 263 respondents, recruited via convenience sampling, completed a series of measures comprising probabilistic reasoning tasks (perception of randomness, base rate, probability, and conjunction fallacy), the Reality Testing subscale of the Inventory of Personality Organization (IPO-RT), the Domain-Specific Risk-Taking Scale, and a terrorism-related behavior change scale. Structural equation modeling examined three progressive models. Firstly, the Independence Model assumed that probabilistic reasoning, perception of risk and reality testing independently predicted terrorism-related behavior change. Secondly, the Mediation Model supposed that probabilistic reasoning and reality testing correlated, and indirectly predicted terrorism-related behavior change through perception of risk. Lastly, the Dual-Influence Model proposed that probabilistic reasoning indirectly predicted terrorism-related behavior change via perception of risk, independent of reality testing. Results indicated that performance on probabilistic reasoning tasks most strongly predicted perception of risk, and preference for an intuitive thinking style (measured by the IPO-RT) best explained terrorism-related behavior change. The combination of perception of risk with probabilistic reasoning ability in the Dual-Influence Model enhanced the predictive power of the analytical-rational route, with conjunction fallacy having a significant indirect effect on terrorism-related behavior change via perception of risk. The Dual-Influence Model possessed superior fit and reported similar predictive relations between intuitive-experiential and analytical-rational routes and terrorism-related behavior change. The discussion critically examines these findings in relation to dual-processing frameworks. This includes considering the limitations of current operationalisations and recommendations for future research that align outcomes and subsequent work more closely to specific dual-process models.

Perception of Risk and Terrorism-Related Behavior Change: Dual Influences of Probabilistic Reasoning and Reality Testing

PubMed Central

Denovan, Andrew; Dagnall, Neil; Drinkwater, Kenneth; Parker, Andrew; Clough, Peter

2017-01-01

The present study assessed the degree to which probabilistic reasoning performance and thinking style influenced perception of risk and self-reported levels of terrorism-related behavior change. A sample of 263 respondents, recruited via convenience sampling, completed a series of measures comprising probabilistic reasoning tasks (perception of randomness, base rate, probability, and conjunction fallacy), the Reality Testing subscale of the Inventory of Personality Organization (IPO-RT), the Domain-Specific Risk-Taking Scale, and a terrorism-related behavior change scale. Structural equation modeling examined three progressive models. Firstly, the Independence Model assumed that probabilistic reasoning, perception of risk and reality testing independently predicted terrorism-related behavior change. Secondly, the Mediation Model supposed that probabilistic reasoning and reality testing correlated, and indirectly predicted terrorism-related behavior change through perception of risk. Lastly, the Dual-Influence Model proposed that probabilistic reasoning indirectly predicted terrorism-related behavior change via perception of risk, independent of reality testing. Results indicated that performance on probabilistic reasoning tasks most strongly predicted perception of risk, and preference for an intuitive thinking style (measured by the IPO-RT) best explained terrorism-related behavior change. The combination of perception of risk with probabilistic reasoning ability in the Dual-Influence Model enhanced the predictive power of the analytical-rational route, with conjunction fallacy having a significant indirect effect on terrorism-related behavior change via perception of risk. The Dual-Influence Model possessed superior fit and reported similar predictive relations between intuitive-experiential and analytical-rational routes and terrorism-related behavior change. The discussion critically examines these findings in relation to dual-processing frameworks. This includes considering the limitations of current operationalisations and recommendations for future research that align outcomes and subsequent work more closely to specific dual-process models. PMID:29062288

Probabilistic Structural Analysis Theory Development

NASA Technical Reports Server (NTRS)

Burnside, O. H.

1985-01-01

The objective of the Probabilistic Structural Analysis Methods (PSAM) project is to develop analysis techniques and computer programs for predicting the probabilistic response of critical structural components for current and future space propulsion systems. This technology will play a central role in establishing system performance and durability. The first year's technical activity is concentrating on probabilistic finite element formulation strategy and code development. Work is also in progress to survey critical materials and space shuttle mian engine components. The probabilistic finite element computer program NESSUS (Numerical Evaluation of Stochastic Structures Under Stress) is being developed. The final probabilistic code will have, in the general case, the capability of performing nonlinear dynamic of stochastic structures. It is the goal of the approximate methods effort to increase problem solving efficiency relative to finite element methods by using energy methods to generate trial solutions which satisfy the structural boundary conditions. These approximate methods will be less computer intensive relative to the finite element approach.

Dual Roles for Spike Signaling in Cortical Neural Populations

PubMed Central

Ballard, Dana H.; Jehee, Janneke F. M.

2011-01-01

A prominent feature of signaling in cortical neurons is that of randomness in the action potential. The output of a typical pyramidal cell can be well fit with a Poisson model, and variations in the Poisson rate repeatedly have been shown to be correlated with stimuli. However while the rate provides a very useful characterization of neural spike data, it may not be the most fundamental description of the signaling code. Recent data showing γ frequency range multi-cell action potential correlations, together with spike timing dependent plasticity, are spurring a re-examination of the classical model, since precise timing codes imply that the generation of spikes is essentially deterministic. Could the observed Poisson randomness and timing determinism reflect two separate modes of communication, or do they somehow derive from a single process? We investigate in a timing-based model whether the apparent incompatibility between these probabilistic and deterministic observations may be resolved by examining how spikes could be used in the underlying neural circuits. The crucial component of this model draws on dual roles for spike signaling. In learning receptive fields from ensembles of inputs, spikes need to behave probabilistically, whereas for fast signaling of individual stimuli, the spikes need to behave deterministically. Our simulations show that this combination is possible if deterministic signals using γ latency coding are probabilistically routed through different members of a cortical cell population at different times. This model exhibits standard features characteristic of Poisson models such as orientation tuning and exponential interval histograms. In addition, it makes testable predictions that follow from the γ latency coding. PMID:21687798

Identification of probabilities.

PubMed

Vitányi, Paul M B; Chater, Nick

2017-02-01

Within psychology, neuroscience and artificial intelligence, there has been increasing interest in the proposal that the brain builds probabilistic models of sensory and linguistic input: that is, to infer a probabilistic model from a sample. The practical problems of such inference are substantial: the brain has limited data and restricted computational resources. But there is a more fundamental question: is the problem of inferring a probabilistic model from a sample possible even in principle? We explore this question and find some surprisingly positive and general results. First, for a broad class of probability distributions characterized by computability restrictions, we specify a learning algorithm that will almost surely identify a probability distribution in the limit given a finite i.i.d. sample of sufficient but unknown length. This is similarly shown to hold for sequences generated by a broad class of Markov chains, subject to computability assumptions. The technical tool is the strong law of large numbers. Second, for a large class of dependent sequences, we specify an algorithm which identifies in the limit a computable measure for which the sequence is typical, in the sense of Martin-Löf (there may be more than one such measure). The technical tool is the theory of Kolmogorov complexity. We analyze the associated predictions in both cases. We also briefly consider special cases, including language learning, and wider theoretical implications for psychology.

Evaluating bacterial gene-finding HMM structures as probabilistic logic programs.

PubMed

Mørk, Søren; Holmes, Ian

2012-03-01

Probabilistic logic programming offers a powerful way to describe and evaluate structured statistical models. To investigate the practicality of probabilistic logic programming for structure learning in bioinformatics, we undertook a simplified bacterial gene-finding benchmark in PRISM, a probabilistic dialect of Prolog. We evaluate Hidden Markov Model structures for bacterial protein-coding gene potential, including a simple null model structure, three structures based on existing bacterial gene finders and two novel model structures. We test standard versions as well as ADPH length modeling and three-state versions of the five model structures. The models are all represented as probabilistic logic programs and evaluated using the PRISM machine learning system in terms of statistical information criteria and gene-finding prediction accuracy, in two bacterial genomes. Neither of our implementations of the two currently most used model structures are best performing in terms of statistical information criteria or prediction performances, suggesting that better-fitting models might be achievable. The source code of all PRISM models, data and additional scripts are freely available for download at: http://github.com/somork/codonhmm. Supplementary data are available at Bioinformatics online.

Integration of Advanced Probabilistic Analysis Techniques with Multi-Physics Models

DOE Office of Scientific and Technical Information (OSTI.GOV)

Cetiner, Mustafa Sacit; none,; Flanagan, George F.

2014-07-30

An integrated simulation platform that couples probabilistic analysis-based tools with model-based simulation tools can provide valuable insights for reactive and proactive responses to plant operating conditions. The objective of this work is to demonstrate the benefits of a partial implementation of the Small Modular Reactor (SMR) Probabilistic Risk Assessment (PRA) Detailed Framework Specification through the coupling of advanced PRA capabilities and accurate multi-physics plant models. Coupling a probabilistic model with a multi-physics model will aid in design, operations, and safety by providing a more accurate understanding of plant behavior. This represents the first attempt at actually integrating these two typesmore » of analyses for a control system used for operations, on a faster than real-time basis. This report documents the development of the basic communication capability to exchange data with the probabilistic model using Reliability Workbench (RWB) and the multi-physics model using Dymola. The communication pathways from injecting a fault (i.e., failing a component) to the probabilistic and multi-physics models were successfully completed. This first version was tested with prototypic models represented in both RWB and Modelica. First, a simple event tree/fault tree (ET/FT) model was created to develop the software code to implement the communication capabilities between the dynamic-link library (dll) and RWB. A program, written in C#, successfully communicates faults to the probabilistic model through the dll. A systems model of the Advanced Liquid-Metal Reactor–Power Reactor Inherently Safe Module (ALMR-PRISM) design developed under another DOE project was upgraded using Dymola to include proper interfaces to allow data exchange with the control application (ConApp). A program, written in C+, successfully communicates faults to the multi-physics model. The results of the example simulation were successfully plotted.« less

Probabilistic structural analysis methods of hot engine structures

NASA Technical Reports Server (NTRS)

Chamis, C. C.; Hopkins, D. A.

1989-01-01

Development of probabilistic structural analysis methods for hot engine structures is a major activity at Lewis Research Center. Recent activities have focused on extending the methods to include the combined uncertainties in several factors on structural response. This paper briefly describes recent progress on composite load spectra models, probabilistic finite element structural analysis, and probabilistic strength degradation modeling. Progress is described in terms of fundamental concepts, computer code development, and representative numerical results.

Developing and Implementing the Data Mining Algorithms in RAVEN

DOE Office of Scientific and Technical Information (OSTI.GOV)

Sen, Ramazan Sonat; Maljovec, Daniel Patrick; Alfonsi, Andrea

The RAVEN code is becoming a comprehensive tool to perform probabilistic risk assessment, uncertainty quantification, and verification and validation. The RAVEN code is being developed to support many programs and to provide a set of methodologies and algorithms for advanced analysis. Scientific computer codes can generate enormous amounts of data. To post-process and analyze such data might, in some cases, take longer than the initial software runtime. Data mining algorithms/methods help in recognizing and understanding patterns in the data, and thus discover knowledge in databases. The methodologies used in the dynamic probabilistic risk assessment or in uncertainty and error quantificationmore » analysis couple system/physics codes with simulation controller codes, such as RAVEN. RAVEN introduces both deterministic and stochastic elements into the simulation while the system/physics code model the dynamics deterministically. A typical analysis is performed by sampling values of a set of parameter values. A major challenge in using dynamic probabilistic risk assessment or uncertainty and error quantification analysis for a complex system is to analyze the large number of scenarios generated. Data mining techniques are typically used to better organize and understand data, i.e. recognizing patterns in the data. This report focuses on development and implementation of Application Programming Interfaces (APIs) for different data mining algorithms, and the application of these algorithms to different databases.« less

Ensemble assimilation of ARGO temperature profile, sea surface temperature, and altimetric satellite data into an eddy permitting primitive equation model of the North Atlantic Ocean

NASA Astrophysics Data System (ADS)

Yan, Y.; Barth, A.; Beckers, J. M.; Candille, G.; Brankart, J. M.; Brasseur, P.

2015-07-01

Sea surface height, sea surface temperature, and temperature profiles at depth collected between January and December 2005 are assimilated into a realistic eddy permitting primitive equation model of the North Atlantic Ocean using the Ensemble Kalman Filter. Sixty ensemble members are generated by adding realistic noise to the forcing parameters related to the temperature. The ensemble is diagnosed and validated by comparison between the ensemble spread and the model/observation difference, as well as by rank histogram before the assimilation experiments. An incremental analysis update scheme is applied in order to reduce spurious oscillations due to the model state correction. The results of the assimilation are assessed according to both deterministic and probabilistic metrics with independent/semiindependent observations. For deterministic validation, the ensemble means, together with the ensemble spreads are compared to the observations, in order to diagnose the ensemble distribution properties in a deterministic way. For probabilistic validation, the continuous ranked probability score (CRPS) is used to evaluate the ensemble forecast system according to reliability and resolution. The reliability is further decomposed into bias and dispersion by the reduced centered random variable (RCRV) score in order to investigate the reliability properties of the ensemble forecast system. The improvement of the assimilation is demonstrated using these validation metrics. Finally, the deterministic validation and the probabilistic validation are analyzed jointly. The consistency and complementarity between both validations are highlighted.

BUMPER: the Bayesian User-friendly Model for Palaeo-Environmental Reconstruction

NASA Astrophysics Data System (ADS)

Holden, Phil; Birks, John; Brooks, Steve; Bush, Mark; Hwang, Grace; Matthews-Bird, Frazer; Valencia, Bryan; van Woesik, Robert

2017-04-01

We describe the Bayesian User-friendly Model for Palaeo-Environmental Reconstruction (BUMPER), a Bayesian transfer function for inferring past climate and other environmental variables from microfossil assemblages. The principal motivation for a Bayesian approach is that the palaeoenvironment is treated probabilistically, and can be updated as additional data become available. Bayesian approaches therefore provide a reconstruction-specific quantification of the uncertainty in the data and in the model parameters. BUMPER is fully self-calibrating, straightforward to apply, and computationally fast, requiring 2 seconds to build a 100-taxon model from a 100-site training-set on a standard personal computer. We apply the model's probabilistic framework to generate thousands of artificial training-sets under ideal assumptions. We then use these to demonstrate both the general applicability of the model and the sensitivity of reconstructions to the characteristics of the training-set, considering assemblage richness, taxon tolerances, and the number of training sites. We demonstrate general applicability to real data, considering three different organism types (chironomids, diatoms, pollen) and different reconstructed variables. In all of these applications an identically configured model is used, the only change being the input files that provide the training-set environment and taxon-count data.

Propagation of the velocity model uncertainties to the seismic event location

NASA Astrophysics Data System (ADS)

Gesret, A.; Desassis, N.; Noble, M.; Romary, T.; Maisons, C.

2015-01-01

Earthquake hypocentre locations are crucial in many domains of application (academic and industrial) as seismic event location maps are commonly used to delineate faults or fractures. The interpretation of these maps depends on location accuracy and on the reliability of the associated uncertainties. The largest contribution to location and uncertainty errors is due to the fact that the velocity model errors are usually not correctly taken into account. We propose a new Bayesian formulation that integrates properly the knowledge on the velocity model into the formulation of the probabilistic earthquake location. In this work, the velocity model uncertainties are first estimated with a Bayesian tomography of active shot data. We implement a sampling Monte Carlo type algorithm to generate velocity models distributed according to the posterior distribution. In a second step, we propagate the velocity model uncertainties to the seismic event location in a probabilistic framework. This enables to obtain more reliable hypocentre locations as well as their associated uncertainties accounting for picking and velocity model uncertainties. We illustrate the tomography results and the gain in accuracy of earthquake location for two synthetic examples and one real data case study in the context of induced microseismicity.

Application of Probabilistic Analysis to Aircraft Impact Dynamics

NASA Technical Reports Server (NTRS)

Lyle, Karen H.; Padula, Sharon L.; Stockwell, Alan E.

2003-01-01

Full-scale aircraft crash simulations performed with nonlinear, transient dynamic, finite element codes can incorporate structural complexities such as: geometrically accurate models; human occupant models; and advanced material models to include nonlinear stressstrain behaviors, laminated composites, and material failure. Validation of these crash simulations is difficult due to a lack of sufficient information to adequately determine the uncertainty in the experimental data and the appropriateness of modeling assumptions. This paper evaluates probabilistic approaches to quantify the uncertainty in the simulated responses. Several criteria are used to determine that a response surface method is the most appropriate probabilistic approach. The work is extended to compare optimization results with and without probabilistic constraints.

Modeling urban air pollution with optimized hierarchical fuzzy inference system.

PubMed

Tashayo, Behnam; Alimohammadi, Abbas

2016-10-01

Environmental exposure assessments (EEA) and epidemiological studies require urban air pollution models with appropriate spatial and temporal resolutions. Uncertain available data and inflexible models can limit air pollution modeling techniques, particularly in under developing countries. This paper develops a hierarchical fuzzy inference system (HFIS) to model air pollution under different land use, transportation, and meteorological conditions. To improve performance, the system treats the issue as a large-scale and high-dimensional problem and develops the proposed model using a three-step approach. In the first step, a geospatial information system (GIS) and probabilistic methods are used to preprocess the data. In the second step, a hierarchical structure is generated based on the problem. In the third step, the accuracy and complexity of the model are simultaneously optimized with a multiple objective particle swarm optimization (MOPSO) algorithm. We examine the capabilities of the proposed model for predicting daily and annual mean PM2.5 and NO2 and compare the accuracy of the results with representative models from existing literature. The benefits provided by the model features, including probabilistic preprocessing, multi-objective optimization, and hierarchical structure, are precisely evaluated by comparing five different consecutive models in terms of accuracy and complexity criteria. Fivefold cross validation is used to assess the performance of the generated models. The respective average RMSEs and coefficients of determination (R (2)) for the test datasets using proposed model are as follows: daily PM2.5 = (8.13, 0.78), annual mean PM2.5 = (4.96, 0.80), daily NO2 = (5.63, 0.79), and annual mean NO2 = (2.89, 0.83). The obtained results demonstrate that the developed hierarchical fuzzy inference system can be utilized for modeling air pollution in EEA and epidemiological studies.

Fracture Mechanics Method for Word Embedding Generation of Neural Probabilistic Linguistic Model.

PubMed

Bi, Size; Liang, Xiao; Huang, Ting-Lei

2016-01-01

Word embedding, a lexical vector representation generated via the neural linguistic model (NLM), is empirically demonstrated to be appropriate for improvement of the performance of traditional language model. However, the supreme dimensionality that is inherent in NLM contributes to the problems of hyperparameters and long-time training in modeling. Here, we propose a force-directed method to improve such problems for simplifying the generation of word embedding. In this framework, each word is assumed as a point in the real world; thus it can approximately simulate the physical movement following certain mechanics. To simulate the variation of meaning in phrases, we use the fracture mechanics to do the formation and breakdown of meaning combined by a 2-gram word group. With the experiments on the natural linguistic tasks of part-of-speech tagging, named entity recognition and semantic role labeling, the result demonstrated that the 2-dimensional word embedding can rival the word embeddings generated by classic NLMs, in terms of accuracy, recall, and text visualization.

Functional model of biological neural networks.

PubMed

Lo, James Ting-Ho

2010-12-01

A functional model of biological neural networks, called temporal hierarchical probabilistic associative memory (THPAM), is proposed in this paper. THPAM comprises functional models of dendritic trees for encoding inputs to neurons, a first type of neuron for generating spike trains, a second type of neuron for generating graded signals to modulate neurons of the first type, supervised and unsupervised Hebbian learning mechanisms for easy learning and retrieving, an arrangement of dendritic trees for maximizing generalization, hardwiring for rotation-translation-scaling invariance, and feedback connections with different delay durations for neurons to make full use of present and past informations generated by neurons in the same and higher layers. These functional models and their processing operations have many functions of biological neural networks that have not been achieved by other models in the open literature and provide logically coherent answers to many long-standing neuroscientific questions. However, biological justifications of these functional models and their processing operations are required for THPAM to qualify as a macroscopic model (or low-order approximate) of biological neural networks.

Qualitative and quantitative comparison of geostatistical techniques of porosity prediction from the seismic and logging data: a case study from the Blackfoot Field, Alberta, Canada

NASA Astrophysics Data System (ADS)

Maurya, S. P.; Singh, K. H.; Singh, N. P.

2018-05-01

In present study, three recently developed geostatistical methods, single attribute analysis, multi-attribute analysis and probabilistic neural network algorithm have been used to predict porosity in inter well region for Blackfoot field, Alberta, Canada, an offshore oil field. These techniques make use of seismic attributes, generated by model based inversion and colored inversion techniques. The principle objective of the study is to find the suitable combination of seismic inversion and geostatistical techniques to predict porosity and identification of prospective zones in 3D seismic volume. The porosity estimated from these geostatistical approaches is corroborated with the well log porosity. The results suggest that all the three implemented geostatistical methods are efficient and reliable to predict the porosity but the multi-attribute and probabilistic neural network analysis provide more accurate and high resolution porosity sections. A low impedance (6000-8000 m/s g/cc) and high porosity (> 15%) zone is interpreted from inverted impedance and porosity sections respectively between 1060 and 1075 ms time interval and is characterized as reservoir. The qualitative and quantitative results demonstrate that of all the employed geostatistical methods, the probabilistic neural network along with model based inversion is the most efficient method for predicting porosity in inter well region.

Extracting and Converting Quantitative Data into Human Error Probabilities

DOE Office of Scientific and Technical Information (OSTI.GOV)

Tuan Q. Tran; Ronald L. Boring; Jeffrey C. Joe

2007-08-01

This paper discusses a proposed method using a combination of advanced statistical approaches (e.g., meta-analysis, regression, structural equation modeling) that will not only convert different empirical results into a common metric for scaling individual PSFs effects, but will also examine the complex interrelationships among PSFs. Furthermore, the paper discusses how the derived statistical estimates (i.e., effect sizes) can be mapped onto a HRA method (e.g. SPAR-H) to generate HEPs that can then be use in probabilistic risk assessment (PRA). The paper concludes with a discussion of the benefits of using academic literature in assisting HRA analysts in generating sound HEPsmore » and HRA developers in validating current HRA models and formulating new HRA models.« less

Tsunamis: stochastic models of occurrence and generation mechanisms

USGS Publications Warehouse

Geist, Eric L.; Oglesby, David D.

2014-01-01

The devastating consequences of the 2004 Indian Ocean and 2011 Japan tsunamis have led to increased research into many different aspects of the tsunami phenomenon. In this entry, we review research related to the observed complexity and uncertainty associated with tsunami generation, propagation, and occurrence described and analyzed using a variety of stochastic methods. In each case, seismogenic tsunamis are primarily considered. Stochastic models are developed from the physical theories that govern tsunami evolution combined with empirical models fitted to seismic and tsunami observations, as well as tsunami catalogs. These stochastic methods are key to providing probabilistic forecasts and hazard assessments for tsunamis. The stochastic methods described here are similar to those described for earthquakes (Vere-Jones 2013) and volcanoes (Bebbington 2013) in this encyclopedia.

Analysis of the stochastic excitability in the flow chemical reactor

DOE Office of Scientific and Technical Information (OSTI.GOV)

Bashkirtseva, Irina

2015-11-30

A dynamic model of the thermochemical process in the flow reactor is considered. We study an influence of the random disturbances on the stationary regime of this model. A phenomenon of noise-induced excitability is demonstrated. For the analysis of this phenomenon, a constructive technique based on the stochastic sensitivity functions and confidence domains is applied. It is shown how elaborated technique can be used for the probabilistic analysis of the generation of mixed-mode stochastic oscillations in the flow chemical reactor.

Mathematical modeling in biological populations through branching processes. Application to salmonid populations.

PubMed

Molina, Manuel; Mota, Manuel; Ramos, Alfonso

2015-01-01

This work deals with mathematical modeling through branching processes. We consider sexually reproducing animal populations where, in each generation, the number of progenitor couples is determined in a non-predictable environment. By using a class of two-sex branching processes, we describe their demographic dynamics and provide several probabilistic and inferential contributions. They include results about the extinction of the population and the estimation of the offspring distribution and its main moments. We also present an application to salmonid populations.

Mathematical Models of Diffusion-Limited Gas Bubble Evolution in Perfused Tissue

DTIC Science & Technology

2013-08-01

the Generation of New Bubbles,” Undersea Biomedical Research, Vol. 18, No. 4 (1991), pp. 333-345. 10. H. D. Van Liew and M. E. Burkard, “Density of...and R. D. Vann, “Probabilistic Gas and Bubble Dynamics Models of Decompression Sickness Occurrence in Air and Nitrogen-Oxygen Diving,” Undersea and...Gas Bubbles During Decompression,” Undersea and Hyperbaric Medicine, Vol. 23, No. 3 (1996), pp. 131-140. 13. R. L. Riley and A. Cournand, “’Ideal

Analysis of the stochastic excitability in the flow chemical reactor

NASA Astrophysics Data System (ADS)

Bashkirtseva, Irina

2015-11-01

A dynamic model of the thermochemical process in the flow reactor is considered. We study an influence of the random disturbances on the stationary regime of this model. A phenomenon of noise-induced excitability is demonstrated. For the analysis of this phenomenon, a constructive technique based on the stochastic sensitivity functions and confidence domains is applied. It is shown how elaborated technique can be used for the probabilistic analysis of the generation of mixed-mode stochastic oscillations in the flow chemical reactor.

Probabilistic Structural Analysis Methods (PSAM) for select space propulsion system structural components

NASA Technical Reports Server (NTRS)

Cruse, T. A.

1987-01-01

The objective is the development of several modular structural analysis packages capable of predicting the probabilistic response distribution for key structural variables such as maximum stress, natural frequencies, transient response, etc. The structural analysis packages are to include stochastic modeling of loads, material properties, geometry (tolerances), and boundary conditions. The solution is to be in terms of the cumulative probability of exceedance distribution (CDF) and confidence bounds. Two methods of probability modeling are to be included as well as three types of structural models - probabilistic finite-element method (PFEM); probabilistic approximate analysis methods (PAAM); and probabilistic boundary element methods (PBEM). The purpose in doing probabilistic structural analysis is to provide the designer with a more realistic ability to assess the importance of uncertainty in the response of a high performance structure. Probabilistic Structural Analysis Method (PSAM) tools will estimate structural safety and reliability, while providing the engineer with information on the confidence that should be given to the predicted behavior. Perhaps most critically, the PSAM results will directly provide information on the sensitivity of the design response to those variables which are seen to be uncertain.

Probabilistic Structural Analysis Methods for select space propulsion system structural components (PSAM)

NASA Technical Reports Server (NTRS)

Cruse, T. A.; Burnside, O. H.; Wu, Y.-T.; Polch, E. Z.; Dias, J. B.

1988-01-01

The objective is the development of several modular structural analysis packages capable of predicting the probabilistic response distribution for key structural variables such as maximum stress, natural frequencies, transient response, etc. The structural analysis packages are to include stochastic modeling of loads, material properties, geometry (tolerances), and boundary conditions. The solution is to be in terms of the cumulative probability of exceedance distribution (CDF) and confidence bounds. Two methods of probability modeling are to be included as well as three types of structural models - probabilistic finite-element method (PFEM); probabilistic approximate analysis methods (PAAM); and probabilistic boundary element methods (PBEM). The purpose in doing probabilistic structural analysis is to provide the designer with a more realistic ability to assess the importance of uncertainty in the response of a high performance structure. Probabilistic Structural Analysis Method (PSAM) tools will estimate structural safety and reliability, while providing the engineer with information on the confidence that should be given to the predicted behavior. Perhaps most critically, the PSAM results will directly provide information on the sensitivity of the design response to those variables which are seen to be uncertain.

Probabilistic machine learning and artificial intelligence.

PubMed

Ghahramani, Zoubin

2015-05-28

How can a machine learn from experience? Probabilistic modelling provides a framework for understanding what learning is, and has therefore emerged as one of the principal theoretical and practical approaches for designing machines that learn from data acquired through experience. The probabilistic framework, which describes how to represent and manipulate uncertainty about models and predictions, has a central role in scientific data analysis, machine learning, robotics, cognitive science and artificial intelligence. This Review provides an introduction to this framework, and discusses some of the state-of-the-art advances in the field, namely, probabilistic programming, Bayesian optimization, data compression and automatic model discovery.

Probabilistic machine learning and artificial intelligence

NASA Astrophysics Data System (ADS)

Ghahramani, Zoubin

2015-05-01

How can a machine learn from experience? Probabilistic modelling provides a framework for understanding what learning is, and has therefore emerged as one of the principal theoretical and practical approaches for designing machines that learn from data acquired through experience. The probabilistic framework, which describes how to represent and manipulate uncertainty about models and predictions, has a central role in scientific data analysis, machine learning, robotics, cognitive science and artificial intelligence. This Review provides an introduction to this framework, and discusses some of the state-of-the-art advances in the field, namely, probabilistic programming, Bayesian optimization, data compression and automatic model discovery.

An investigation into the probabilistic combination of quasi-static and random accelerations

NASA Technical Reports Server (NTRS)

Schock, R. W.; Tuell, L. P.

1984-01-01

The development of design load factors for aerospace and aircraft components and experiment support structures, which are subject to a simultaneous vehicle dynamic vibration (quasi-static) and acoustically generated random vibration, require the selection of a combination methodology. Typically, the procedure is to define the quasi-static and the random generated response separately, and arithmetically add or root sum square to get combined accelerations. Since the combination of a probabilistic and a deterministic function yield a probabilistic function, a viable alternate approach would be to determine the characteristics of the combined acceleration probability density function and select an appropriate percentile level for the combined acceleration. The following paper develops this mechanism and provides graphical data to select combined accelerations for most popular percentile levels.

A global empirical system for probabilistic seasonal climate prediction

NASA Astrophysics Data System (ADS)

Eden, J. M.; van Oldenborgh, G. J.; Hawkins, E.; Suckling, E. B.

2015-12-01

Preparing for episodes with risks of anomalous weather a month to a year ahead is an important challenge for governments, non-governmental organisations, and private companies and is dependent on the availability of reliable forecasts. The majority of operational seasonal forecasts are made using process-based dynamical models, which are complex, computationally challenging and prone to biases. Empirical forecast approaches built on statistical models to represent physical processes offer an alternative to dynamical systems and can provide either a benchmark for comparison or independent supplementary forecasts. Here, we present a simple empirical system based on multiple linear regression for producing probabilistic forecasts of seasonal surface air temperature and precipitation across the globe. The global CO2-equivalent concentration is taken as the primary predictor; subsequent predictors, including large-scale modes of variability in the climate system and local-scale information, are selected on the basis of their physical relationship with the predictand. The focus given to the climate change signal as a source of skill and the probabilistic nature of the forecasts produced constitute a novel approach to global empirical prediction. Hindcasts for the period 1961-2013 are validated against observations using deterministic (correlation of seasonal means) and probabilistic (continuous rank probability skill scores) metrics. Good skill is found in many regions, particularly for surface air temperature and most notably in much of Europe during the spring and summer seasons. For precipitation, skill is generally limited to regions with known El Niño-Southern Oscillation (ENSO) teleconnections. The system is used in a quasi-operational framework to generate empirical seasonal forecasts on a monthly basis.

An empirical system for probabilistic seasonal climate prediction

NASA Astrophysics Data System (ADS)

Eden, Jonathan; van Oldenborgh, Geert Jan; Hawkins, Ed; Suckling, Emma

2016-04-01

Preparing for episodes with risks of anomalous weather a month to a year ahead is an important challenge for governments, non-governmental organisations, and private companies and is dependent on the availability of reliable forecasts. The majority of operational seasonal forecasts are made using process-based dynamical models, which are complex, computationally challenging and prone to biases. Empirical forecast approaches built on statistical models to represent physical processes offer an alternative to dynamical systems and can provide either a benchmark for comparison or independent supplementary forecasts. Here, we present a simple empirical system based on multiple linear regression for producing probabilistic forecasts of seasonal surface air temperature and precipitation across the globe. The global CO2-equivalent concentration is taken as the primary predictor; subsequent predictors, including large-scale modes of variability in the climate system and local-scale information, are selected on the basis of their physical relationship with the predictand. The focus given to the climate change signal as a source of skill and the probabilistic nature of the forecasts produced constitute a novel approach to global empirical prediction. Hindcasts for the period 1961-2013 are validated against observations using deterministic (correlation of seasonal means) and probabilistic (continuous rank probability skill scores) metrics. Good skill is found in many regions, particularly for surface air temperature and most notably in much of Europe during the spring and summer seasons. For precipitation, skill is generally limited to regions with known El Niño-Southern Oscillation (ENSO) teleconnections. The system is used in a quasi-operational framework to generate empirical seasonal forecasts on a monthly basis.

Process for computing geometric perturbations for probabilistic analysis

DOEpatents

Fitch, Simeon H. K. [Charlottesville, VA; Riha, David S [San Antonio, TX; Thacker, Ben H [San Antonio, TX

2012-04-10

A method for computing geometric perturbations for probabilistic analysis. The probabilistic analysis is based on finite element modeling, in which uncertainties in the modeled system are represented by changes in the nominal geometry of the model, referred to as "perturbations". These changes are accomplished using displacement vectors, which are computed for each node of a region of interest and are based on mean-value coordinate calculations.

Probabilistic modeling of discourse-aware sentence processing.

PubMed

Dubey, Amit; Keller, Frank; Sturt, Patrick

2013-07-01

Probabilistic models of sentence comprehension are increasingly relevant to questions concerning human language processing. However, such models are often limited to syntactic factors. This restriction is unrealistic in light of experimental results suggesting interactions between syntax and other forms of linguistic information in human sentence processing. To address this limitation, this article introduces two sentence processing models that augment a syntactic component with information about discourse co-reference. The novel combination of probabilistic syntactic components with co-reference classifiers permits them to more closely mimic human behavior than existing models. The first model uses a deep model of linguistics, based in part on probabilistic logic, allowing it to make qualitative predictions on experimental data; the second model uses shallow processing to make quantitative predictions on a broad-coverage reading-time corpus. Copyright © 2013 Cognitive Science Society, Inc.

Seasonal streamflow prediction using ensemble streamflow prediction technique for the Rangitata and Waitaki River basins on the South Island of New Zealand

NASA Astrophysics Data System (ADS)

Singh, Shailesh Kumar

2014-05-01

Streamflow forecasts are essential for making critical decision for optimal allocation of water supplies for various demands that include irrigation for agriculture, habitat for fisheries, hydropower production and flood warning. The major objective of this study is to explore the Ensemble Streamflow Prediction (ESP) based forecast in New Zealand catchments and to highlights the present capability of seasonal flow forecasting of National Institute of Water and Atmospheric Research (NIWA). In this study a probabilistic forecast framework for ESP is presented. The basic assumption in ESP is that future weather pattern were experienced historically. Hence, past forcing data can be used with current initial condition to generate an ensemble of prediction. Small differences in initial conditions can result in large difference in the forecast. The initial state of catchment can be obtained by continuously running the model till current time and use this initial state with past forcing data to generate ensemble of flow for future. The approach taken here is to run TopNet hydrological models with a range of past forcing data (precipitation, temperature etc.) with current initial conditions. The collection of runs is called the ensemble. ESP give probabilistic forecasts for flow. From ensemble members the probability distributions can be derived. The probability distributions capture part of the intrinsic uncertainty in weather or climate. An ensemble stream flow prediction which provide probabilistic hydrological forecast with lead time up to 3 months is presented for Rangitata, Ahuriri, and Hooker and Jollie rivers in South Island of New Zealand. ESP based seasonal forecast have better skill than climatology. This system can provide better over all information for holistic water resource management.

Probabilistic calibration of the distributed hydrological model RIBS applied to real-time flood forecasting: the Harod river basin case study (Israel)

NASA Astrophysics Data System (ADS)

Nesti, Alice; Mediero, Luis; Garrote, Luis; Caporali, Enrica

2010-05-01

An automatic probabilistic calibration method for distributed rainfall-runoff models is presented. The high number of parameters in hydrologic distributed models makes special demands on the optimization procedure to estimate model parameters. With the proposed technique it is possible to reduce the complexity of calibration while maintaining adequate model predictions. The first step of the calibration procedure of the main model parameters is done manually with the aim to identify their variation range. Afterwards a Monte-Carlo technique is applied, which consists on repetitive model simulations with randomly generated parameters. The Monte Carlo Analysis Toolbox (MCAT) includes a number of analysis methods to evaluate the results of these Monte Carlo parameter sampling experiments. The study investigates the use of a global sensitivity analysis as a screening tool to reduce the parametric dimensionality of multi-objective hydrological model calibration problems, while maximizing the information extracted from hydrological response data. The method is applied to the calibration of the RIBS flood forecasting model in the Harod river basin, placed on Israel. The Harod basin has an extension of 180 km2. The catchment has a Mediterranean climate and it is mainly characterized by a desert landscape, with a soil that is able to absorb large quantities of rainfall and at the same time is capable to generate high peaks of discharge. Radar rainfall data with 6 minute temporal resolution are available as input to the model. The aim of the study is the validation of the model for real-time flood forecasting, in order to evaluate the benefits of improved precipitation forecasting within the FLASH European project.

A random spatial network model based on elementary postulates

USGS Publications Warehouse

Karlinger, Michael R.; Troutman, Brent M.

1989-01-01

A model for generating random spatial networks that is based on elementary postulates comparable to those of the random topology model is proposed. In contrast to the random topology model, this model ascribes a unique spatial specification to generated drainage networks, a distinguishing property of some network growth models. The simplicity of the postulates creates an opportunity for potential analytic investigations of the probabilistic structure of the drainage networks, while the spatial specification enables analyses of spatially dependent network properties. In the random topology model all drainage networks, conditioned on magnitude (number of first-order streams), are equally likely, whereas in this model all spanning trees of a grid, conditioned on area and drainage density, are equally likely. As a result, link lengths in the generated networks are not independent, as usually assumed in the random topology model. For a preliminary model evaluation, scale-dependent network characteristics, such as geometric diameter and link length properties, and topologic characteristics, such as bifurcation ratio, are computed for sets of drainage networks generated on square and rectangular grids. Statistics of the bifurcation and length ratios fall within the range of values reported for natural drainage networks, but geometric diameters tend to be relatively longer than those for natural networks.

Developing EHR-driven heart failure risk prediction models using CPXR(Log) with the probabilistic loss function.

PubMed

Taslimitehrani, Vahid; Dong, Guozhu; Pereira, Naveen L; Panahiazar, Maryam; Pathak, Jyotishman

2016-04-01

Computerized survival prediction in healthcare identifying the risk of disease mortality, helps healthcare providers to effectively manage their patients by providing appropriate treatment options. In this study, we propose to apply a classification algorithm, Contrast Pattern Aided Logistic Regression (CPXR(Log)) with the probabilistic loss function, to develop and validate prognostic risk models to predict 1, 2, and 5year survival in heart failure (HF) using data from electronic health records (EHRs) at Mayo Clinic. The CPXR(Log) constructs a pattern aided logistic regression model defined by several patterns and corresponding local logistic regression models. One of the models generated by CPXR(Log) achieved an AUC and accuracy of 0.94 and 0.91, respectively, and significantly outperformed prognostic models reported in prior studies. Data extracted from EHRs allowed incorporation of patient co-morbidities into our models which helped improve the performance of the CPXR(Log) models (15.9% AUC improvement), although did not improve the accuracy of the models built by other classifiers. We also propose a probabilistic loss function to determine the large error and small error instances. The new loss function used in the algorithm outperforms other functions used in the previous studies by 1% improvement in the AUC. This study revealed that using EHR data to build prediction models can be very challenging using existing classification methods due to the high dimensionality and complexity of EHR data. The risk models developed by CPXR(Log) also reveal that HF is a highly heterogeneous disease, i.e., different subgroups of HF patients require different types of considerations with their diagnosis and treatment. Our risk models provided two valuable insights for application of predictive modeling techniques in biomedicine: Logistic risk models often make systematic prediction errors, and it is prudent to use subgroup based prediction models such as those given by CPXR(Log) when investigating heterogeneous diseases. Copyright © 2016 Elsevier Inc. All rights reserved.

Tsunamigenic scenarios for southern Peru and northern Chile seismic gap: Deterministic and probabilistic hybrid approach for hazard assessment

NASA Astrophysics Data System (ADS)

González-Carrasco, J. F.; Gonzalez, G.; Aránguiz, R.; Yanez, G. A.; Melgar, D.; Salazar, P.; Shrivastava, M. N.; Das, R.; Catalan, P. A.; Cienfuegos, R.

2017-12-01

Plausible worst-case tsunamigenic scenarios definition plays a relevant role in tsunami hazard assessment focused in emergency preparedness and evacuation planning for coastal communities. During the last decade, the occurrence of major and moderate tsunamigenic earthquakes along worldwide subduction zones has given clues about critical parameters involved in near-field tsunami inundation processes, i.e. slip spatial distribution, shelf resonance of edge waves and local geomorphology effects. To analyze the effects of these seismic and hydrodynamic variables over the epistemic uncertainty of coastal inundation, we implement a combined methodology using deterministic and probabilistic approaches to construct 420 tsunamigenic scenarios in a mature seismic gap of southern Peru and northern Chile, extended from 17ºS to 24ºS. The deterministic scenarios are calculated using a regional distribution of trench-parallel gravity anomaly (TPGA) and trench-parallel topography anomaly (TPTA), three-dimensional Slab 1.0 worldwide subduction zones geometry model and published interseismic coupling (ISC) distributions. As result, we find four higher slip deficit zones interpreted as major seismic asperities of the gap, used in a hierarchical tree scheme to generate ten tsunamigenic scenarios with seismic magnitudes fluctuates between Mw 8.4 to Mw 8.9. Additionally, we construct ten homogeneous slip scenarios as inundation baseline. For the probabilistic approach, we implement a Karhunen - Loève expansion to generate 400 stochastic tsunamigenic scenarios over the maximum extension of the gap, with the same magnitude range of the deterministic sources. All the scenarios are simulated through a non-hydrostatic tsunami model Neowave 2D, using a classical nesting scheme, for five coastal major cities in northern Chile (Arica, Iquique, Tocopilla, Mejillones and Antofagasta) obtaining high resolution data of inundation depth, runup, coastal currents and sea level elevation. The probabilistic kinematic tsunamigenic scenarios give a more realistic slip patterns, similar to maximum slip amount of major past earthquakes. For all studied sites, the peak of slip location and shelf resonance is a first order control for the observed coastal inundation depths results.

A semi-supervised learning approach for RNA secondary structure prediction.

PubMed

Yonemoto, Haruka; Asai, Kiyoshi; Hamada, Michiaki

2015-08-01

RNA secondary structure prediction is a key technology in RNA bioinformatics. Most algorithms for RNA secondary structure prediction use probabilistic models, in which the model parameters are trained with reliable RNA secondary structures. Because of the difficulty of determining RNA secondary structures by experimental procedures, such as NMR or X-ray crystal structural analyses, there are still many RNA sequences that could be useful for training whose secondary structures have not been experimentally determined. In this paper, we introduce a novel semi-supervised learning approach for training parameters in a probabilistic model of RNA secondary structures in which we employ not only RNA sequences with annotated secondary structures but also ones with unknown secondary structures. Our model is based on a hybrid of generative (stochastic context-free grammars) and discriminative models (conditional random fields) that has been successfully applied to natural language processing. Computational experiments indicate that the accuracy of secondary structure prediction is improved by incorporating RNA sequences with unknown secondary structures into training. To our knowledge, this is the first study of a semi-supervised learning approach for RNA secondary structure prediction. This technique will be useful when the number of reliable structures is limited. Copyright © 2015 Elsevier Ltd. All rights reserved.

A probabilistic NF2 relational algebra for integrated information retrieval and database systems

DOE Office of Scientific and Technical Information (OSTI.GOV)

Fuhr, N.; Roelleke, T.

The integration of information retrieval (IR) and database systems requires a data model which allows for modelling documents as entities, representing uncertainty and vagueness and performing uncertain inference. For this purpose, we present a probabilistic data model based on relations in non-first-normal-form (NF2). Here, tuples are assigned probabilistic weights giving the probability that a tuple belongs to a relation. Thus, the set of weighted index terms of a document are represented as a probabilistic subrelation. In a similar way, imprecise attribute values are modelled as a set-valued attribute. We redefine the relational operators for this type of relations such thatmore » the result of each operator is again a probabilistic NF2 relation, where the weight of a tuple gives the probability that this tuple belongs to the result. By ordering the tuples according to decreasing probabilities, the model yields a ranking of answers like in most IR models. This effect also can be used for typical database queries involving imprecise attribute values as well as for combinations of database and IR queries.« less

Neural networks and fault probability evaluation for diagnosis issues.

PubMed

Kourd, Yahia; Lefebvre, Dimitri; Guersi, Noureddine

2014-01-01

This paper presents a new FDI technique for fault detection and isolation in unknown nonlinear systems. The objective of the research is to construct and analyze residuals by means of artificial intelligence and probabilistic methods. Artificial neural networks are first used for modeling issues. Neural networks models are designed for learning the fault-free and the faulty behaviors of the considered systems. Once the residuals generated, an evaluation using probabilistic criteria is applied to them to determine what is the most likely fault among a set of candidate faults. The study also includes a comparison between the contributions of these tools and their limitations, particularly through the establishment of quantitative indicators to assess their performance. According to the computation of a confidence factor, the proposed method is suitable to evaluate the reliability of the FDI decision. The approach is applied to detect and isolate 19 fault candidates in the DAMADICS benchmark. The results obtained with the proposed scheme are compared with the results obtained according to a usual thresholding method.

Decoding brain activity using a large-scale probabilistic functional-anatomical atlas of human cognition

PubMed Central

Jones, Michael N.

2017-01-01

A central goal of cognitive neuroscience is to decode human brain activity—that is, to infer mental processes from observed patterns of whole-brain activation. Previous decoding efforts have focused on classifying brain activity into a small set of discrete cognitive states. To attain maximal utility, a decoding framework must be open-ended, systematic, and context-sensitive—that is, capable of interpreting numerous brain states, presented in arbitrary combinations, in light of prior information. Here we take steps towards this objective by introducing a probabilistic decoding framework based on a novel topic model—Generalized Correspondence Latent Dirichlet Allocation—that learns latent topics from a database of over 11,000 published fMRI studies. The model produces highly interpretable, spatially-circumscribed topics that enable flexible decoding of whole-brain images. Importantly, the Bayesian nature of the model allows one to “seed” decoder priors with arbitrary images and text—enabling researchers, for the first time, to generate quantitative, context-sensitive interpretations of whole-brain patterns of brain activity. PMID:29059185

An alternative approach to probabilistic seismic hazard analysis in the Aegean region using Monte Carlo simulation

NASA Astrophysics Data System (ADS)

Weatherill, Graeme; Burton, Paul W.

2010-09-01

The Aegean is the most seismically active and tectonically complex region in Europe. Damaging earthquakes have occurred here throughout recorded history, often resulting in considerable loss of life. The Monte Carlo method of probabilistic seismic hazard analysis (PSHA) is used to determine the level of ground motion likely to be exceeded in a given time period. Multiple random simulations of seismicity are generated to calculate, directly, the ground motion for a given site. Within the seismic hazard analysis we explore the impact of different seismic source models, incorporating both uniform zones and distributed seismicity. A new, simplified, seismic source model, derived from seismotectonic interpretation, is presented for the Aegean region. This is combined into the epistemic uncertainty analysis alongside existing source models for the region, and models derived by a K-means cluster analysis approach. Seismic source models derived using the K-means approach offer a degree of objectivity and reproducibility into the otherwise subjective approach of delineating seismic sources using expert judgment. Similar review and analysis is undertaken for the selection of peak ground acceleration (PGA) attenuation models, incorporating into the epistemic analysis Greek-specific models, European models and a Next Generation Attenuation model. Hazard maps for PGA on a "rock" site with a 10% probability of being exceeded in 50 years are produced and different source and attenuation models are compared. These indicate that Greek-specific attenuation models, with their smaller aleatory variability terms, produce lower PGA hazard, whilst recent European models and Next Generation Attenuation (NGA) model produce similar results. The Monte Carlo method is extended further to assimilate epistemic uncertainty into the hazard calculation, thus integrating across several appropriate source and PGA attenuation models. Site condition and fault-type are also integrated into the hazard mapping calculations. These hazard maps are in general agreement with previous maps for the Aegean, recognising the highest hazard in the Ionian Islands, Gulf of Corinth and Hellenic Arc. Peak Ground Accelerations for some sites in these regions reach as high as 500-600 cm s -2 using European/NGA attenuation models, and 400-500 cm s -2 using Greek attenuation models.

Incorporating networks in a probabilistic graphical model to find drivers for complex human diseases.

PubMed

Mezlini, Aziz M; Goldenberg, Anna

2017-10-01

Discovering genetic mechanisms driving complex diseases is a hard problem. Existing methods often lack power to identify the set of responsible genes. Protein-protein interaction networks have been shown to boost power when detecting gene-disease associations. We introduce a Bayesian framework, Conflux, to find disease associated genes from exome sequencing data using networks as a prior. There are two main advantages to using networks within a probabilistic graphical model. First, networks are noisy and incomplete, a substantial impediment to gene discovery. Incorporating networks into the structure of a probabilistic models for gene inference has less impact on the solution than relying on the noisy network structure directly. Second, using a Bayesian framework we can keep track of the uncertainty of each gene being associated with the phenotype rather than returning a fixed list of genes. We first show that using networks clearly improves gene detection compared to individual gene testing. We then show consistently improved performance of Conflux compared to the state-of-the-art diffusion network-based method Hotnet2 and a variety of other network and variant aggregation methods, using randomly generated and literature-reported gene sets. We test Hotnet2 and Conflux on several network configurations to reveal biases and patterns of false positives and false negatives in each case. Our experiments show that our novel Bayesian framework Conflux incorporates many of the advantages of the current state-of-the-art methods, while offering more flexibility and improved power in many gene-disease association scenarios.

A Probabilistic Model for Predicting Attenuation of Viruses During Percolation in Unsaturated Natural Barriers

NASA Astrophysics Data System (ADS)

Faulkner, B. R.; Lyon, W. G.

2001-12-01

We present a probabilistic model for predicting virus attenuation. The solution employs the assumption of complete mixing. Monte Carlo methods are used to generate ensemble simulations of virus attenuation due to physical, biological, and chemical factors. The model generates a probability of failure to achieve 4-log attenuation. We tabulated data from related studies to develop probability density functions for input parameters, and utilized a database of soil hydraulic parameters based on the 12 USDA soil categories. Regulators can use the model based on limited information such as boring logs, climate data, and soil survey reports for a particular site of interest. Plackett-Burman sensitivity analysis indicated the most important main effects on probability of failure to achieve 4-log attenuation in our model were mean logarithm of saturated hydraulic conductivity (+0.396), mean water content (+0.203), mean solid-water mass transfer coefficient (-0.147), and the mean solid-water equilibrium partitioning coefficient (-0.144). Using the model, we predicted the probability of failure of a one-meter thick proposed hydrogeologic barrier and a water content of 0.3. With the currently available data and the associated uncertainty, we predicted soils classified as sand would fail (p=0.999), silt loams would also fail (p=0.292), but soils classified as clays would provide the required 4-log attenuation (p=0.001). The model is extendible in the sense that probability density functions of parameters can be modified as future studies refine the uncertainty, and the lightweight object-oriented design of the computer model (implemented in Java) will facilitate reuse with modified classes. This is an abstract of a proposed presentation and does not necessarily reflect EPA policy.

Estimating the Expected Value of Sample Information Using the Probabilistic Sensitivity Analysis Sample

PubMed Central

Oakley, Jeremy E.; Brennan, Alan; Breeze, Penny

2015-01-01

Health economic decision-analytic models are used to estimate the expected net benefits of competing decision options. The true values of the input parameters of such models are rarely known with certainty, and it is often useful to quantify the value to the decision maker of reducing uncertainty through collecting new data. In the context of a particular decision problem, the value of a proposed research design can be quantified by its expected value of sample information (EVSI). EVSI is commonly estimated via a 2-level Monte Carlo procedure in which plausible data sets are generated in an outer loop, and then, conditional on these, the parameters of the decision model are updated via Bayes rule and sampled in an inner loop. At each iteration of the inner loop, the decision model is evaluated. This is computationally demanding and may be difficult if the posterior distribution of the model parameters conditional on sampled data is hard to sample from. We describe a fast nonparametric regression-based method for estimating per-patient EVSI that requires only the probabilistic sensitivity analysis sample (i.e., the set of samples drawn from the joint distribution of the parameters and the corresponding net benefits). The method avoids the need to sample from the posterior distributions of the parameters and avoids the need to rerun the model. The only requirement is that sample data sets can be generated. The method is applicable with a model of any complexity and with any specification of model parameter distribution. We demonstrate in a case study the superior efficiency of the regression method over the 2-level Monte Carlo method. PMID:25810269

Scientific assessment of accuracy, skill and reliability of ocean probabilistic forecast products.

NASA Astrophysics Data System (ADS)

Wei, M.; Rowley, C. D.; Barron, C. N.; Hogan, P. J.

2016-02-01

As ocean operational centers are increasingly adopting and generating probabilistic forecast products for their customers with valuable forecast uncertainties, how to assess and measure these complicated probabilistic forecast products objectively is challenging. The first challenge is how to deal with the huge amount of the data from the ensemble forecasts. The second one is how to describe the scientific quality of probabilistic products. In fact, probabilistic forecast accuracy, skills, reliability, resolutions are different attributes of a forecast system. We briefly introduce some of the fundamental metrics such as the Reliability Diagram, Reliability, Resolution, Brier Score (BS), Brier Skill Score (BSS), Ranked Probability Score (RPS), Ranked Probability Skill Score (RPSS), Continuous Ranked Probability Score (CRPS), and Continuous Ranked Probability Skill Score (CRPSS). The values and significance of these metrics are demonstrated for the forecasts from the US Navy's regional ensemble system with different ensemble members. The advantages and differences of these metrics are studied and clarified.

Reliability Analysis and Standardization of Spacecraft Command Generation Processes

NASA Technical Reports Server (NTRS)

Meshkat, Leila; Grenander, Sven; Evensen, Ken

2011-01-01

center dot In order to reduce commanding errors that are caused by humans, we create an approach and corresponding artifacts for standardizing the command generation process and conducting risk management during the design and assurance of such processes. center dot The literature review conducted during the standardization process revealed that very few atomic level human activities are associated with even a broad set of missions. center dot Applicable human reliability metrics for performing these atomic level tasks are available. center dot The process for building a "Periodic Table" of Command and Control Functions as well as Probabilistic Risk Assessment (PRA) models is demonstrated. center dot The PRA models are executed using data from human reliability data banks. center dot The Periodic Table is related to the PRA models via Fault Links.

Solar thermoelectricity via advanced latent heat storage: A cost-effective small-scale CSP application

NASA Astrophysics Data System (ADS)

Glatzmaier, G. C.; Rea, J.; Olsen, M. L.; Oshman, C.; Hardin, C.; Alleman, J.; Sharp, J.; Weigand, R.; Campo, D.; Hoeschele, G.; Parilla, P. A.; Siegel, N. P.; Toberer, E. S.; Ginley, D. S.

2017-06-01

We are developing a novel concentrating solar electricity-generating technology that is both modular and dispatchable. Solar ThermoElectricity via Advanced Latent heat Storage (STEALS) uses concentrated solar flux to generate high-temperature thermal energy, which directly converts to electricity via thermoelectric generators (TEGs), stored within a phase-change material (PCM) for electricity generation at a later time, or both allowing for simultaneous charging of the PCM and electricity generation. STEALS has inherent features that drive its cost-competitive scale to be much smaller than current commercial concentrating solar power (CSP) plants. Most obvious is modularity of the solid-state TEG, which favors smaller scales in the kilowatt range as compared to CSP steam turbines, which are minimally 50 MWe for commercial power plants. Here, we present techno-economic and market analyses that show STEALS can be a cost-effective electricity-generating technology with particular appeal to small-scale microgrid applications. We evaluated levelized cost of energy (LCOE) for STEALS and for a comparable photovoltaic (PV) system with battery storage. For STEALS, we estimated capital costs and the LCOE as functions of the type of PCM including the use of recycled aluminum alloys, and evaluated the cost tradeoffs between plasma spray coatings and solution-based boron coatings that are applied to the wetted surfaces of the PCM subsystem. We developed a probabilistic cost model that accounts for uncertainties in the cost and performance inputs to the LCOE estimation. Our probabilistic model estimated LCOE for a 100-kWe STEALS system that had 5 hours of thermal storage and 8-10 hours of total daily power generation. For these cases, the solar multiple for the heliostat field varied between 1.12 and 1.5. We identified microgrids as a likely market for the STEALS system. We characterized microgrid markets in terms of nominal power, dispatchability, geographic location, and customer type, and specified additional features for STEALS that are needed to meet the needs of this growing power market.

A Probabilistic Model of Local Sequence Alignment That Simplifies Statistical Significance Estimation

PubMed Central

Eddy, Sean R.

2008-01-01

Sequence database searches require accurate estimation of the statistical significance of scores. Optimal local sequence alignment scores follow Gumbel distributions, but determining an important parameter of the distribution (λ) requires time-consuming computational simulation. Moreover, optimal alignment scores are less powerful than probabilistic scores that integrate over alignment uncertainty (“Forward” scores), but the expected distribution of Forward scores remains unknown. Here, I conjecture that both expected score distributions have simple, predictable forms when full probabilistic modeling methods are used. For a probabilistic model of local sequence alignment, optimal alignment bit scores (“Viterbi” scores) are Gumbel-distributed with constant λ = log 2, and the high scoring tail of Forward scores is exponential with the same constant λ. Simulation studies support these conjectures over a wide range of profile/sequence comparisons, using 9,318 profile-hidden Markov models from the Pfam database. This enables efficient and accurate determination of expectation values (E-values) for both Viterbi and Forward scores for probabilistic local alignments. PMID:18516236

Award-Winning CARES/Life Ceramics Durability Evaluation Software Is Making Advanced Technology Accessible

NASA Technical Reports Server (NTRS)

1997-01-01

Products made from advanced ceramics show great promise for revolutionizing aerospace and terrestrial propulsion and power generation. However, ceramic components are difficult to design because brittle materials in general have widely varying strength values. The CARES/Life software developed at the NASA Lewis Research Center eases this by providing a tool that uses probabilistic reliability analysis techniques to optimize the design and manufacture of brittle material components. CARES/Life is an integrated package that predicts the probability of a monolithic ceramic component's failure as a function of its time in service. It couples commercial finite element programs--which resolve a component's temperature and stress distribution - with reliability evaluation and fracture mechanics routines for modeling strength - limiting defects. These routines are based on calculations of the probabilistic nature of the brittle material's strength.

Fuzzy-probabilistic model for risk assessment of radioactive material railway transportation.

PubMed

Avramenko, M; Bolyatko, V; Kosterev, V

2005-01-01

Transportation of radioactive materials is obviously accompanied by a certain risk. A model for risk assessment of emergency situations and terrorist attacks may be useful for choosing possible routes and for comparing the various defence strategies. In particular, risk assessment is crucial for safe transportation of excess weapons-grade plutonium arising from the removal of plutonium from military employment. A fuzzy-probabilistic model for risk assessment of railway transportation has been developed taking into account the different natures of risk-affecting parameters (probabilistic and not probabilistic but fuzzy). Fuzzy set theory methods as well as standard methods of probability theory have been used for quantitative risk assessment. Information-preserving transformations are applied to realise the correct aggregation of probabilistic and fuzzy parameters. Estimations have also been made of the inhalation doses resulting from possible accidents during plutonium transportation. The obtained data show the scale of possible consequences that may arise from plutonium transportation accidents.

Upgrades to the REA method for producing probabilistic climate change projections

NASA Astrophysics Data System (ADS)

Xu, Ying; Gao, Xuejie; Giorgi, Filippo

2010-05-01

We present an augmented version of the Reliability Ensemble Averaging (REA) method designed to generate probabilistic climate change information from ensembles of climate model simulations. Compared to the original version, the augmented one includes consideration of multiple variables and statistics in the calculation of the performance-based weights. In addition, the model convergence criterion previously employed is removed. The method is applied to the calculation of changes in mean and variability for temperature and precipitation over different sub-regions of East Asia based on the recently completed CMIP3 multi-model ensemble. Comparison of the new and old REA methods, along with the simple averaging procedure, and the use of different combinations of performance metrics shows that at fine sub-regional scales the choice of weighting is relevant. This is mostly because the models show a substantial spread in performance for the simulation of precipitation statistics, a result that supports the use of model weighting as a useful option to account for wide ranges of quality of models. The REA method, and in particular the upgraded one, provides a simple and flexible framework for assessing the uncertainty related to the aggregation of results from ensembles of models in order to produce climate change information at the regional scale. KEY WORDS: REA method, Climate change, CMIP3

What do we gain with Probabilistic Flood Loss Models?

NASA Astrophysics Data System (ADS)

Schroeter, K.; Kreibich, H.; Vogel, K.; Merz, B.; Lüdtke, S.

2015-12-01

The reliability of flood loss models is a prerequisite for their practical usefulness. Oftentimes, traditional uni-variate damage models as for instance depth-damage curves fail to reproduce the variability of observed flood damage. Innovative multi-variate probabilistic modelling approaches are promising to capture and quantify the uncertainty involved and thus to improve the basis for decision making. In this study we compare the predictive capability of two probabilistic modelling approaches, namely Bagging Decision Trees and Bayesian Networks and traditional stage damage functions which are cast in a probabilistic framework. For model evaluation we use empirical damage data which are available from computer aided telephone interviews that were respectively compiled after the floods in 2002, 2005, 2006 and 2013 in the Elbe and Danube catchments in Germany. We carry out a split sample test by sub-setting the damage records. One sub-set is used to derive the models and the remaining records are used to evaluate the predictive performance of the model. Further we stratify the sample according to catchments which allows studying model performance in a spatial transfer context. Flood damage estimation is carried out on the scale of the individual buildings in terms of relative damage. The predictive performance of the models is assessed in terms of systematic deviations (mean bias), precision (mean absolute error) as well as in terms of reliability which is represented by the proportion of the number of observations that fall within the 95-quantile and 5-quantile predictive interval. The reliability of the probabilistic predictions within validation runs decreases only slightly and achieves a very good coverage of observations within the predictive interval. Probabilistic models provide quantitative information about prediction uncertainty which is crucial to assess the reliability of model predictions and improves the usefulness of model results.

A comprehensive probabilistic analysis model of oil pipelines network based on Bayesian network

NASA Astrophysics Data System (ADS)

Zhang, C.; Qin, T. X.; Jiang, B.; Huang, C.

2018-02-01

Oil pipelines network is one of the most important facilities of energy transportation. But oil pipelines network accident may result in serious disasters. Some analysis models for these accidents have been established mainly based on three methods, including event-tree, accident simulation and Bayesian network. Among these methods, Bayesian network is suitable for probabilistic analysis. But not all the important influencing factors are considered and the deployment rule of the factors has not been established. This paper proposed a probabilistic analysis model of oil pipelines network based on Bayesian network. Most of the important influencing factors, including the key environment condition and emergency response are considered in this model. Moreover, the paper also introduces a deployment rule for these factors. The model can be used in probabilistic analysis and sensitive analysis of oil pipelines network accident.

Computation of marginal distributions of peak-heights in electropherograms for analysing single source and mixture STR DNA samples.

PubMed

Cowell, Robert G

2018-05-04

Current models for single source and mixture samples, and probabilistic genotyping software based on them used for analysing STR electropherogram data, assume simple probability distributions, such as the gamma distribution, to model the allelic peak height variability given the initial amount of DNA prior to PCR amplification. Here we illustrate how amplicon number distributions, for a model of the process of sample DNA collection and PCR amplification, may be efficiently computed by evaluating probability generating functions using discrete Fourier transforms. Copyright © 2018 Elsevier B.V. All rights reserved.

Unbiased, scalable sampling of protein loop conformations from probabilistic priors.

PubMed

Zhang, Yajia; Hauser, Kris

2013-01-01

Protein loops are flexible structures that are intimately tied to function, but understanding loop motion and generating loop conformation ensembles remain significant computational challenges. Discrete search techniques scale poorly to large loops, optimization and molecular dynamics techniques are prone to local minima, and inverse kinematics techniques can only incorporate structural preferences in adhoc fashion. This paper presents Sub-Loop Inverse Kinematics Monte Carlo (SLIKMC), a new Markov chain Monte Carlo algorithm for generating conformations of closed loops according to experimentally available, heterogeneous structural preferences. Our simulation experiments demonstrate that the method computes high-scoring conformations of large loops (>10 residues) orders of magnitude faster than standard Monte Carlo and discrete search techniques. Two new developments contribute to the scalability of the new method. First, structural preferences are specified via a probabilistic graphical model (PGM) that links conformation variables, spatial variables (e.g., atom positions), constraints and prior information in a unified framework. The method uses a sparse PGM that exploits locality of interactions between atoms and residues. Second, a novel method for sampling sub-loops is developed to generate statistically unbiased samples of probability densities restricted by loop-closure constraints. Numerical experiments confirm that SLIKMC generates conformation ensembles that are statistically consistent with specified structural preferences. Protein conformations with 100+ residues are sampled on standard PC hardware in seconds. Application to proteins involved in ion-binding demonstrate its potential as a tool for loop ensemble generation and missing structure completion.

Unbiased, scalable sampling of protein loop conformations from probabilistic priors

PubMed Central

2013-01-01

Background Protein loops are flexible structures that are intimately tied to function, but understanding loop motion and generating loop conformation ensembles remain significant computational challenges. Discrete search techniques scale poorly to large loops, optimization and molecular dynamics techniques are prone to local minima, and inverse kinematics techniques can only incorporate structural preferences in adhoc fashion. This paper presents Sub-Loop Inverse Kinematics Monte Carlo (SLIKMC), a new Markov chain Monte Carlo algorithm for generating conformations of closed loops according to experimentally available, heterogeneous structural preferences. Results Our simulation experiments demonstrate that the method computes high-scoring conformations of large loops (>10 residues) orders of magnitude faster than standard Monte Carlo and discrete search techniques. Two new developments contribute to the scalability of the new method. First, structural preferences are specified via a probabilistic graphical model (PGM) that links conformation variables, spatial variables (e.g., atom positions), constraints and prior information in a unified framework. The method uses a sparse PGM that exploits locality of interactions between atoms and residues. Second, a novel method for sampling sub-loops is developed to generate statistically unbiased samples of probability densities restricted by loop-closure constraints. Conclusion Numerical experiments confirm that SLIKMC generates conformation ensembles that are statistically consistent with specified structural preferences. Protein conformations with 100+ residues are sampled on standard PC hardware in seconds. Application to proteins involved in ion-binding demonstrate its potential as a tool for loop ensemble generation and missing structure completion. PMID:24565175

Accuracy of latent-variable estimation in Bayesian semi-supervised learning.

PubMed

Yamazaki, Keisuke

2015-09-01

Hierarchical probabilistic models, such as Gaussian mixture models, are widely used for unsupervised learning tasks. These models consist of observable and latent variables, which represent the observable data and the underlying data-generation process, respectively. Unsupervised learning tasks, such as cluster analysis, are regarded as estimations of latent variables based on the observable ones. The estimation of latent variables in semi-supervised learning, where some labels are observed, will be more precise than that in unsupervised, and one of the concerns is to clarify the effect of the labeled data. However, there has not been sufficient theoretical analysis of the accuracy of the estimation of latent variables. In a previous study, a distribution-based error function was formulated, and its asymptotic form was calculated for unsupervised learning with generative models. It has been shown that, for the estimation of latent variables, the Bayes method is more accurate than the maximum-likelihood method. The present paper reveals the asymptotic forms of the error function in Bayesian semi-supervised learning for both discriminative and generative models. The results show that the generative model, which uses all of the given data, performs better when the model is well specified. Copyright © 2015 Elsevier Ltd. All rights reserved.

Bayesian Monitoring Systems for the CTBT: Historical Development and New Results

NASA Astrophysics Data System (ADS)

Russell, S.; Arora, N. S.; Moore, D.

2016-12-01

A project at Berkeley, begun in 2009 in collaboration with CTBTO andmore recently with LLNL, has reformulated the global seismicmonitoring problem in a Bayesian framework. A first-generation system,NETVISA, has been built comprising a spatial event prior andgenerative models of event transmission and detection, as well as aMonte Carlo inference algorithm. The probabilistic model allows forseamless integration of various disparate sources of information,including negative information (the absence of detections). Workingfrom arrivals extracted by traditional station processing fromInternational Monitoring System (IMS) data, NETVISA achieves areduction of around 60% in the number of missed events compared withthe currently deployed network processing system. It also finds manyevents that are missed by the human analysts who postprocess the IMSoutput. Recent improvements include the integration of models forinfrasound and hydroacoustic detections and a global depth model fornatural seismicity trained from ISC data. NETVISA is now fullycompatible with the CTBTO operating environment. A second-generation model called SIGVISA extends NETVISA's generativemodel all the way from events to raw signal data, avoiding theerror-prone bottom-up detection phase of station processing. SIGVISA'smodel automatically captures the phenomena underlying existingdetection and location techniques such as multilateration, waveformcorrelation matching, and double-differencing, and integrates theminto a global inference process that also (like NETVISA) handles denovo events. Initial results for the Western US in early 2008 (whenthe transportable US Array was operating) shows that SIGVISA finds,from IMS data only, more than twice the number of events recorded inthe CTBTO Late Event Bulletin (LEB). For mb 1.0-2.5, the ratio is more than10; put another way, for this data set, SIGVISA lowers the detectionthreshold by roughly one magnitude compared to LEB. The broader message of this work is that probabilistic inference basedon a vertically integrated generative model that directly expressesgeophysical knowledge can be a much more effective approach forinterpreting scientific data than the traditional bottom-up processingpipeline.

Using Bayesian Nonparametric Hidden Semi-Markov Models to Disentangle Affect Processes during Marital Interaction

PubMed Central

Griffin, William A.; Li, Xun

2016-01-01

Sequential affect dynamics generated during the interaction of intimate dyads, such as married couples, are associated with a cascade of effects—some good and some bad—on each partner, close family members, and other social contacts. Although the effects are well documented, the probabilistic structures associated with micro-social processes connected to the varied outcomes remain enigmatic. Using extant data we developed a method of classifying and subsequently generating couple dynamics using a Hierarchical Dirichlet Process Hidden semi-Markov Model (HDP-HSMM). Our findings indicate that several key aspects of existing models of marital interaction are inadequate: affect state emissions and their durations, along with the expected variability differences between distressed and nondistressed couples are present but highly nuanced; and most surprisingly, heterogeneity among highly satisfied couples necessitate that they be divided into subgroups. We review how this unsupervised learning technique generates plausible dyadic sequences that are sensitive to relationship quality and provide a natural mechanism for computational models of behavioral and affective micro-social processes. PMID:27187319

Temporally consistent probabilistic detection of new multiple sclerosis lesions in brain MRI.

PubMed

Elliott, Colm; Arnold, Douglas L; Collins, D Louis; Arbel, Tal

2013-08-01

Detection of new Multiple Sclerosis (MS) lesions on magnetic resonance imaging (MRI) is important as a marker of disease activity and as a potential surrogate for relapses. We propose an approach where sequential scans are jointly segmented, to provide a temporally consistent tissue segmentation while remaining sensitive to newly appearing lesions. The method uses a two-stage classification process: 1) a Bayesian classifier provides a probabilistic brain tissue classification at each voxel of reference and follow-up scans, and 2) a random-forest based lesion-level classification provides a final identification of new lesions. Generative models are learned based on 364 scans from 95 subjects from a multi-center clinical trial. The method is evaluated on sequential brain MRI of 160 subjects from a separate multi-center clinical trial, and is compared to 1) semi-automatically generated ground truth segmentations and 2) fully manual identification of new lesions generated independently by nine expert raters on a subset of 60 subjects. For new lesions greater than 0.15 cc in size, the classifier has near perfect performance (99% sensitivity, 2% false detection rate), as compared to ground truth. The proposed method was also shown to exceed the performance of any one of the nine expert manual identifications.

Exploring a potential energy surface by machine learning for characterizing atomic transport

NASA Astrophysics Data System (ADS)

Kanamori, Kenta; Toyoura, Kazuaki; Honda, Junya; Hattori, Kazuki; Seko, Atsuto; Karasuyama, Masayuki; Shitara, Kazuki; Shiga, Motoki; Kuwabara, Akihide; Takeuchi, Ichiro

2018-03-01

We propose a machine-learning method for evaluating the potential barrier governing atomic transport based on the preferential selection of dominant points for atomic transport. The proposed method generates numerous random samples of the entire potential energy surface (PES) from a probabilistic Gaussian process model of the PES, which enables defining the likelihood of the dominant points. The robustness and efficiency of the method are demonstrated on a dozen model cases for proton diffusion in oxides, in comparison with a conventional nudge elastic band method.

Data free inference with processed data products

DOE PAGES

Chowdhary, K.; Najm, H. N.

2014-07-12

Here, we consider the context of probabilistic inference of model parameters given error bars or confidence intervals on model output values, when the data is unavailable. We introduce a class of algorithms in a Bayesian framework, relying on maximum entropy arguments and approximate Bayesian computation methods, to generate consistent data with the given summary statistics. Once we obtain consistent data sets, we pool the respective posteriors, to arrive at a single, averaged density on the parameters. This approach allows us to perform accurate forward uncertainty propagation consistent with the reported statistics.

Optimizing Negotiation Conflict in the Cloud Service Negotiation Framework Using Probabilistic Decision Making Model

PubMed Central

Rajavel, Rajkumar; Thangarathinam, Mala

2015-01-01

Optimization of negotiation conflict in the cloud service negotiation framework is identified as one of the major challenging issues. This negotiation conflict occurs during the bilateral negotiation process between the participants due to the misperception, aggressive behavior, and uncertain preferences and goals about their opponents. Existing research work focuses on the prerequest context of negotiation conflict optimization by grouping similar negotiation pairs using distance, binary, context-dependent, and fuzzy similarity approaches. For some extent, these approaches can maximize the success rate and minimize the communication overhead among the participants. To further optimize the success rate and communication overhead, the proposed research work introduces a novel probabilistic decision making model for optimizing the negotiation conflict in the long-term negotiation context. This decision model formulates the problem of managing different types of negotiation conflict that occurs during negotiation process as a multistage Markov decision problem. At each stage of negotiation process, the proposed decision model generates the heuristic decision based on the past negotiation state information without causing any break-off among the participants. In addition, this heuristic decision using the stochastic decision tree scenario can maximize the revenue among the participants available in the cloud service negotiation framework. PMID:26543899

Optimizing Negotiation Conflict in the Cloud Service Negotiation Framework Using Probabilistic Decision Making Model.

PubMed

Rajavel, Rajkumar; Thangarathinam, Mala

2015-01-01

Optimization of negotiation conflict in the cloud service negotiation framework is identified as one of the major challenging issues. This negotiation conflict occurs during the bilateral negotiation process between the participants due to the misperception, aggressive behavior, and uncertain preferences and goals about their opponents. Existing research work focuses on the prerequest context of negotiation conflict optimization by grouping similar negotiation pairs using distance, binary, context-dependent, and fuzzy similarity approaches. For some extent, these approaches can maximize the success rate and minimize the communication overhead among the participants. To further optimize the success rate and communication overhead, the proposed research work introduces a novel probabilistic decision making model for optimizing the negotiation conflict in the long-term negotiation context. This decision model formulates the problem of managing different types of negotiation conflict that occurs during negotiation process as a multistage Markov decision problem. At each stage of negotiation process, the proposed decision model generates the heuristic decision based on the past negotiation state information without causing any break-off among the participants. In addition, this heuristic decision using the stochastic decision tree scenario can maximize the revenue among the participants available in the cloud service negotiation framework.

Landslide Hazard Probability Derived from Inherent and Dynamic Determinants

NASA Astrophysics Data System (ADS)

Strauch, Ronda; Istanbulluoglu, Erkan

2016-04-01

Landslide hazard research has typically been conducted independently from hydroclimate research. We unify these two lines of research to provide regional scale landslide hazard information for risk assessments and resource management decision-making. Our approach combines an empirical inherent landslide probability with a numerical dynamic probability, generated by combining routed recharge from the Variable Infiltration Capacity (VIC) macro-scale land surface hydrologic model with a finer resolution probabilistic slope stability model run in a Monte Carlo simulation. Landslide hazard mapping is advanced by adjusting the dynamic model of stability with an empirically-based scalar representing the inherent stability of the landscape, creating a probabilistic quantitative measure of geohazard prediction at a 30-m resolution. Climatology, soil, and topography control the dynamic nature of hillslope stability and the empirical information further improves the discriminating ability of the integrated model. This work will aid resource management decision-making in current and future landscape and climatic conditions. The approach is applied as a case study in North Cascade National Park Complex, a rugged terrain with nearly 2,700 m (9,000 ft) of vertical relief, covering 2757 sq km (1064 sq mi) in northern Washington State, U.S.A.

Goal-directed decision making as probabilistic inference: A computational framework and potential neural correlates

PubMed Central

Solway, A.; Botvinick, M.

2013-01-01

Recent work has given rise to the view that reward-based decision making is governed by two key controllers: a habit system, which stores stimulus-response associations shaped by past reward, and a goal-oriented system that selects actions based on their anticipated outcomes. The current literature provides a rich body of computational theory addressing habit formation, centering on temporal-difference learning mechanisms. Less progress has been made toward formalizing the processes involved in goal-directed decision making. We draw on recent work in cognitive neuroscience, animal conditioning, cognitive and developmental psychology and machine learning, to outline a new theory of goal-directed decision making. Our basic proposal is that the brain, within an identifiable network of cortical and subcortical structures, implements a probabilistic generative model of reward, and that goal-directed decision making is effected through Bayesian inversion of this model. We present a set of simulations implementing the account, which address benchmark behavioral and neuroscientific findings, and which give rise to a set of testable predictions. We also discuss the relationship between the proposed framework and other models of decision making, including recent models of perceptual choice, to which our theory bears a direct connection. PMID:22229491

Hierarchical Probabilistic Inference of Cosmic Shear

NASA Astrophysics Data System (ADS)

Schneider, Michael D.; Hogg, David W.; Marshall, Philip J.; Dawson, William A.; Meyers, Joshua; Bard, Deborah J.; Lang, Dustin

2015-07-01

Point estimators for the shearing of galaxy images induced by gravitational lensing involve a complex inverse problem in the presence of noise, pixelization, and model uncertainties. We present a probabilistic forward modeling approach to gravitational lensing inference that has the potential to mitigate the biased inferences in most common point estimators and is practical for upcoming lensing surveys. The first part of our statistical framework requires specification of a likelihood function for the pixel data in an imaging survey given parameterized models for the galaxies in the images. We derive the lensing shear posterior by marginalizing over all intrinsic galaxy properties that contribute to the pixel data (i.e., not limited to galaxy ellipticities) and learn the distributions for the intrinsic galaxy properties via hierarchical inference with a suitably flexible conditional probabilitiy distribution specification. We use importance sampling to separate the modeling of small imaging areas from the global shear inference, thereby rendering our algorithm computationally tractable for large surveys. With simple numerical examples we demonstrate the improvements in accuracy from our importance sampling approach, as well as the significance of the conditional distribution specification for the intrinsic galaxy properties when the data are generated from an unknown number of distinct galaxy populations with different morphological characteristics.

Probabilistic Fatigue Damage Prognosis Using a Surrogate Model Trained Via 3D Finite Element Analysis

NASA Technical Reports Server (NTRS)

Leser, Patrick E.; Hochhalter, Jacob D.; Newman, John A.; Leser, William P.; Warner, James E.; Wawrzynek, Paul A.; Yuan, Fuh-Gwo

2015-01-01

Utilizing inverse uncertainty quantification techniques, structural health monitoring can be integrated with damage progression models to form probabilistic predictions of a structure's remaining useful life. However, damage evolution in realistic structures is physically complex. Accurately representing this behavior requires high-fidelity models which are typically computationally prohibitive. In the present work, a high-fidelity finite element model is represented by a surrogate model, reducing computation times. The new approach is used with damage diagnosis data to form a probabilistic prediction of remaining useful life for a test specimen under mixed-mode conditions.

Probabilistic models of cognition: conceptual foundations.

PubMed

Chater, Nick; Tenenbaum, Joshua B; Yuille, Alan

2006-07-01

Remarkable progress in the mathematics and computer science of probability has led to a revolution in the scope of probabilistic models. In particular, 'sophisticated' probabilistic methods apply to structured relational systems such as graphs and grammars, of immediate relevance to the cognitive sciences. This Special Issue outlines progress in this rapidly developing field, which provides a potentially unifying perspective across a wide range of domains and levels of explanation. Here, we introduce the historical and conceptual foundations of the approach, explore how the approach relates to studies of explicit probabilistic reasoning, and give a brief overview of the field as it stands today.

Development of Probabilistic Flood Inundation Mapping For Flooding Induced by Dam Failure

NASA Astrophysics Data System (ADS)

Tsai, C.; Yeh, J. J. J.

2017-12-01

A primary function of flood inundation mapping is to forecast flood hazards and assess potential losses. However, uncertainties limit the reliability of inundation hazard assessments. Major sources of uncertainty should be taken into consideration by an optimal flood management strategy. This study focuses on the 20km reach downstream of the Shihmen Reservoir in Taiwan. A dam failure induced flood herein provides the upstream boundary conditions of flood routing. The two major sources of uncertainty that are considered in the hydraulic model and the flood inundation mapping herein are uncertainties in the dam break model and uncertainty of the roughness coefficient. The perturbance moment method is applied to a dam break model and the hydro system model to develop probabilistic flood inundation mapping. Various numbers of uncertain variables can be considered in these models and the variability of outputs can be quantified. The probabilistic flood inundation mapping for dam break induced floods can be developed with consideration of the variability of output using a commonly used HEC-RAS model. Different probabilistic flood inundation mappings are discussed and compared. Probabilistic flood inundation mappings are hoped to provide new physical insights in support of the evaluation of concerning reservoir flooded areas.

Automated segmentation of the prostate in 3D MR images using a probabilistic atlas and a spatially constrained deformable model.

PubMed

Martin, Sébastien; Troccaz, Jocelyne; Daanenc, Vincent

2010-04-01

The authors present a fully automatic algorithm for the segmentation of the prostate in three-dimensional magnetic resonance (MR) images. The approach requires the use of an anatomical atlas which is built by computing transformation fields mapping a set of manually segmented images to a common reference. These transformation fields are then applied to the manually segmented structures of the training set in order to get a probabilistic map on the atlas. The segmentation is then realized through a two stage procedure. In the first stage, the processed image is registered to the probabilistic atlas. Subsequently, a probabilistic segmentation is obtained by mapping the probabilistic map of the atlas to the patient's anatomy. In the second stage, a deformable surface evolves toward the prostate boundaries by merging information coming from the probabilistic segmentation, an image feature model and a statistical shape model. During the evolution of the surface, the probabilistic segmentation allows the introduction of a spatial constraint that prevents the deformable surface from leaking in an unlikely configuration. The proposed method is evaluated on 36 exams that were manually segmented by a single expert. A median Dice similarity coefficient of 0.86 and an average surface error of 2.41 mm are achieved. By merging prior knowledge, the presented method achieves a robust and completely automatic segmentation of the prostate in MR images. Results show that the use of a spatial constraint is useful to increase the robustness of the deformable model comparatively to a deformable surface that is only driven by an image appearance model.

Assessing and reporting uncertainties in dietary exposure analysis: Mapping of uncertainties in a tiered approach.

PubMed

Kettler, Susanne; Kennedy, Marc; McNamara, Cronan; Oberdörfer, Regina; O'Mahony, Cian; Schnabel, Jürgen; Smith, Benjamin; Sprong, Corinne; Faludi, Roland; Tennant, David

2015-08-01

Uncertainty analysis is an important component of dietary exposure assessments in order to understand correctly the strength and limits of its results. Often, standard screening procedures are applied in a first step which results in conservative estimates. If through those screening procedures a potential exceedance of health-based guidance values is indicated, within the tiered approach more refined models are applied. However, the sources and types of uncertainties in deterministic and probabilistic models can vary or differ. A key objective of this work has been the mapping of different sources and types of uncertainties to better understand how to best use uncertainty analysis to generate more realistic comprehension of dietary exposure. In dietary exposure assessments, uncertainties can be introduced by knowledge gaps about the exposure scenario, parameter and the model itself. With this mapping, general and model-independent uncertainties have been identified and described, as well as those which can be introduced and influenced by the specific model during the tiered approach. This analysis identifies that there are general uncertainties common to point estimates (screening or deterministic methods) and probabilistic exposure assessment methods. To provide further clarity, general sources of uncertainty affecting many dietary exposure assessments should be separated from model-specific uncertainties. Copyright © 2015 The Authors. Published by Elsevier Ltd.. All rights reserved.

A Probabilistic Model for Hydrokinetic Turbine Collision Risks: Exploring Impacts on Fish

PubMed Central

Hammar, Linus; Eggertsen, Linda; Andersson, Sandra; Ehnberg, Jimmy; Arvidsson, Rickard; Gullström, Martin; Molander, Sverker

2015-01-01

A variety of hydrokinetic turbines are currently under development for power generation in rivers, tidal straits and ocean currents. Because some of these turbines are large, with rapidly moving rotor blades, the risk of collision with aquatic animals has been brought to attention. The behavior and fate of animals that approach such large hydrokinetic turbines have not yet been monitored at any detail. In this paper, we conduct a synthesis of the current knowledge and understanding of hydrokinetic turbine collision risks. The outcome is a generic fault tree based probabilistic model suitable for estimating population-level ecological risks. New video-based data on fish behavior in strong currents are provided and models describing fish avoidance behaviors are presented. The findings indicate low risk for small-sized fish. However, at large turbines (≥5 m), bigger fish seem to have high probability of collision, mostly because rotor detection and avoidance is difficult in low visibility. Risks can therefore be substantial for vulnerable populations of large-sized fish, which thrive in strong currents. The suggested collision risk model can be applied to different turbine designs and at a variety of locations as basis for case-specific risk assessments. The structure of the model facilitates successive model validation, refinement and application to other organism groups such as marine mammals. PMID:25730314

A probabilistic model for hydrokinetic turbine collision risks: exploring impacts on fish.

PubMed

Hammar, Linus; Eggertsen, Linda; Andersson, Sandra; Ehnberg, Jimmy; Arvidsson, Rickard; Gullström, Martin; Molander, Sverker

2015-01-01

A variety of hydrokinetic turbines are currently under development for power generation in rivers, tidal straits and ocean currents. Because some of these turbines are large, with rapidly moving rotor blades, the risk of collision with aquatic animals has been brought to attention. The behavior and fate of animals that approach such large hydrokinetic turbines have not yet been monitored at any detail. In this paper, we conduct a synthesis of the current knowledge and understanding of hydrokinetic turbine collision risks. The outcome is a generic fault tree based probabilistic model suitable for estimating population-level ecological risks. New video-based data on fish behavior in strong currents are provided and models describing fish avoidance behaviors are presented. The findings indicate low risk for small-sized fish. However, at large turbines (≥5 m), bigger fish seem to have high probability of collision, mostly because rotor detection and avoidance is difficult in low visibility. Risks can therefore be substantial for vulnerable populations of large-sized fish, which thrive in strong currents. The suggested collision risk model can be applied to different turbine designs and at a variety of locations as basis for case-specific risk assessments. The structure of the model facilitates successive model validation, refinement and application to other organism groups such as marine mammals.

Fine-Scale Event Location and Error Analysis in NET-VISA

NASA Astrophysics Data System (ADS)

Arora, N. S.; Russell, S.

2016-12-01

NET-VISA is a generative probabilistic model for the occurrence of seismic, hydro, and atmospheric events, and the propagation of energy from these events through various mediums and phases before being detected, or misdetected, by IMS stations. It is built on top of the basic station, and arrival detection processing at the IDC, and is currently being tested in the IDC network processing pipelines. A key distinguishing feature of NET-VISA is that it is easy to incorporate prior scientific knowledge and historical data into the probabilistic model. The model accounts for both detections and mis-detections when forming events, and this allows it to make more accurate event hypothesis. It has been continuously evaluated since 2012, and in each year it makes a roughly 60% reduction in the number of missed events without increasing the false event rate as compared to the existing GA algorithm. More importantly the model finds large numbers of events that have been confirmed by regional seismic bulletins but missed by the IDC analysts using the same data. In this work we focus on enhancements to the model to improve the location accuracy, and error ellipses. We will present a new version of the model that focuses on the fine scale around the event location, and present error ellipses and analysis of recent important events.

Probabilistic tsunami inundation map based on stochastic earthquake source model: A demonstration case in Macau, the South China Sea

NASA Astrophysics Data System (ADS)

Li, Linlin; Switzer, Adam D.; Wang, Yu; Chan, Chung-Han; Qiu, Qiang; Weiss, Robert

2017-04-01

Current tsunami inundation maps are commonly generated using deterministic scenarios, either for real-time forecasting or based on hypothetical "worst-case" events. Such maps are mainly used for emergency response and evacuation planning and do not include the information of return period. However, in practice, probabilistic tsunami inundation maps are required in a wide variety of applications, such as land-use planning, engineer design and for insurance purposes. In this study, we present a method to develop the probabilistic tsunami inundation map using a stochastic earthquake source model. To demonstrate the methodology, we take Macau a coastal city in the South China Sea as an example. Two major advances of this method are: it incorporates the most updated information of seismic tsunamigenic sources along the Manila megathrust; it integrates a stochastic source model into a Monte Carlo-type simulation in which a broad range of slip distribution patterns are generated for large numbers of synthetic earthquake events. When aggregated the large amount of inundation simulation results, we analyze the uncertainties associated with variability of earthquake rupture location and slip distribution. We also explore how tsunami hazard evolves in Macau in the context of sea level rise. Our results suggest Macau faces moderate tsunami risk due to its low-lying elevation, extensive land reclamation, high coastal population and major infrastructure density. Macau consists of four districts: Macau Peninsula, Taipa Island, Coloane island and Cotai strip. Of these Macau Peninsula is the most vulnerable to tsunami due to its low-elevation and exposure to direct waves and refracted waves from the offshore region and reflected waves from mainland. Earthquakes with magnitude larger than Mw8.0 in the northern Manila trench would likely cause hazardous inundation in Macau. Using a stochastic source model, we are able to derive a spread of potential tsunami impacts for earthquakes with the same magnitude. The diversity is caused by both random rupture locations and heterogeneous slip distribution. Adding the sea level rise component, the inundated depth caused by 1 m sea level rise is equivalent to the one caused by 90 percentile of an ensemble of Mw8.4 earthquakes.

Cost effectiveness of a pharmacist-led information technology intervention for reducing rates of clinically important errors in medicines management in general practices (PINCER).

PubMed

Elliott, Rachel A; Putman, Koen D; Franklin, Matthew; Annemans, Lieven; Verhaeghe, Nick; Eden, Martin; Hayre, Jasdeep; Rodgers, Sarah; Sheikh, Aziz; Avery, Anthony J

2014-06-01

We recently showed that a pharmacist-led information technology-based intervention (PINCER) was significantly more effective in reducing medication errors in general practices than providing simple feedback on errors, with cost per error avoided at £79 (US$131). We aimed to estimate cost effectiveness of the PINCER intervention by combining effectiveness in error reduction and intervention costs with the effect of the individual errors on patient outcomes and healthcare costs, to estimate the effect on costs and QALYs. We developed Markov models for each of six medication errors targeted by PINCER. Clinical event probability, treatment pathway, resource use and costs were extracted from literature and costing tariffs. A composite probabilistic model combined patient-level error models with practice-level error rates and intervention costs from the trial. Cost per extra QALY and cost-effectiveness acceptability curves were generated from the perspective of NHS England, with a 5-year time horizon. The PINCER intervention generated £2,679 less cost and 0.81 more QALYs per practice [incremental cost-effectiveness ratio (ICER): -£3,037 per QALY] in the deterministic analysis. In the probabilistic analysis, PINCER generated 0.001 extra QALYs per practice compared with simple feedback, at £4.20 less per practice. Despite this extremely small set of differences in costs and outcomes, PINCER dominated simple feedback with a mean ICER of -£3,936 (standard error £2,970). At a ceiling 'willingness-to-pay' of £20,000/QALY, PINCER reaches 59 % probability of being cost effective. PINCER produced marginal health gain at slightly reduced overall cost. Results are uncertain due to the poor quality of data to inform the effect of avoiding errors.

Training-Image Based Geostatistical Inversion Using a Spatial Generative Adversarial Neural Network

NASA Astrophysics Data System (ADS)

Laloy, Eric; Hérault, Romain; Jacques, Diederik; Linde, Niklas

2018-01-01

Probabilistic inversion within a multiple-point statistics framework is often computationally prohibitive for high-dimensional problems. To partly address this, we introduce and evaluate a new training-image based inversion approach for complex geologic media. Our approach relies on a deep neural network of the generative adversarial network (GAN) type. After training using a training image (TI), our proposed spatial GAN (SGAN) can quickly generate 2-D and 3-D unconditional realizations. A key characteristic of our SGAN is that it defines a (very) low-dimensional parameterization, thereby allowing for efficient probabilistic inversion using state-of-the-art Markov chain Monte Carlo (MCMC) methods. In addition, available direct conditioning data can be incorporated within the inversion. Several 2-D and 3-D categorical TIs are first used to analyze the performance of our SGAN for unconditional geostatistical simulation. Training our deep network can take several hours. After training, realizations containing a few millions of pixels/voxels can be produced in a matter of seconds. This makes it especially useful for simulating many thousands of realizations (e.g., for MCMC inversion) as the relative cost of the training per realization diminishes with the considered number of realizations. Synthetic inversion case studies involving 2-D steady state flow and 3-D transient hydraulic tomography with and without direct conditioning data are used to illustrate the effectiveness of our proposed SGAN-based inversion. For the 2-D case, the inversion rapidly explores the posterior model distribution. For the 3-D case, the inversion recovers model realizations that fit the data close to the target level and visually resemble the true model well.

A cost-effectiveness analysis of a proactive management strategy for the Sprint Fidelis recall: a probabilistic decision analysis model.

PubMed

Bashir, Jamil; Cowan, Simone; Raymakers, Adam; Yamashita, Michael; Danter, Matthew; Krahn, Andrew; Lynd, Larry D

2013-12-01

The management of the recall is complicated by the competing risks of lead failure and complications that can occur with lead revision. Many of these patients are currently undergoing an elective generator change--an ideal time to consider lead revision. To determine the cost-effectiveness of a proactive management strategy for the Sprint Fidelis recall. We obtained detailed clinical outcomes and costing data from a retrospective analysis of 341 patients who received the Sprint Fidelis lead in British Columbia, where patients younger than 60 years were offered lead extraction when undergoing generator replacement. These population-based data were used to construct and populate a probabilistic Markov model in which a proactive management strategy was compared to a conservative strategy to determine the incremental cost per lead failure avoided. In our population, elective lead revisions were half the cost of emergent revisions and had a lower complication rate. In the model, the incremental cost-effectiveness ratio of proactive lead revision versus a recommended monitoring strategy was $12,779 per lead failure avoided. The proactive strategy resulted in 21 fewer failures per 100 patients treated and reduced the chance of an additional complication from an unexpected surgery. Cost-effectiveness analysis suggests that prospective lead revision should be considered when patients with a Sprint Fidelis lead present for pulse generator change. Elective revision of the lead is justified even when 25% of the population is operated on per year, and in some scenarios, it is both less costly and provides a better outcome. © 2013 Heart Rhythm Society Published by Heart Rhythm Society All rights reserved.

A method to encapsulate model structural uncertainty in ensemble projections of future climate: EPIC v1.0

NASA Astrophysics Data System (ADS)

Lewis, Jared; Bodeker, Greg E.; Kremser, Stefanie; Tait, Andrew

2017-12-01

A method, based on climate pattern scaling, has been developed to expand a small number of projections of fields of a selected climate variable (X) into an ensemble that encapsulates a wide range of indicative model structural uncertainties. The method described in this paper is referred to as the Ensemble Projections Incorporating Climate model uncertainty (EPIC) method. Each ensemble member is constructed by adding contributions from (1) a climatology derived from observations that represents the time-invariant part of the signal; (2) a contribution from forced changes in X, where those changes can be statistically related to changes in global mean surface temperature (Tglobal); and (3) a contribution from unforced variability that is generated by a stochastic weather generator. The patterns of unforced variability are also allowed to respond to changes in Tglobal. The statistical relationships between changes in X (and its patterns of variability) and Tglobal are obtained in a training phase. Then, in an implementation phase, 190 simulations of Tglobal are generated using a simple climate model tuned to emulate 19 different global climate models (GCMs) and 10 different carbon cycle models. Using the generated Tglobal time series and the correlation between the forced changes in X and Tglobal, obtained in the training phase, the forced change in the X field can be generated many times using Monte Carlo analysis. A stochastic weather generator is used to generate realistic representations of weather which include spatial coherence. Because GCMs and regional climate models (RCMs) are less likely to correctly represent unforced variability compared to observations, the stochastic weather generator takes as input measures of variability derived from observations, but also responds to forced changes in climate in a way that is consistent with the RCM projections. This approach to generating a large ensemble of projections is many orders of magnitude more computationally efficient than running multiple GCM or RCM simulations. Such a large ensemble of projections permits a description of a probability density function (PDF) of future climate states rather than a small number of individual story lines within that PDF, which may not be representative of the PDF as a whole; the EPIC method largely corrects for such potential sampling biases. The method is useful for providing projections of changes in climate to users wishing to investigate the impacts and implications of climate change in a probabilistic way. A web-based tool, using the EPIC method to provide probabilistic projections of changes in daily maximum and minimum temperatures for New Zealand, has been developed and is described in this paper.

The Effects of Climate Model Similarity on Local, Risk-Based Adaptation Planning

NASA Astrophysics Data System (ADS)

Steinschneider, S.; Brown, C. M.

2014-12-01

The climate science community has recently proposed techniques to develop probabilistic projections of climate change from ensemble climate model output. These methods provide a means to incorporate the formal concept of risk, i.e., the product of impact and probability, into long-term planning assessments for local systems under climate change. However, approaches for pdf development often assume that different climate models provide independent information for the estimation of probabilities, despite model similarities that stem from a common genealogy. Here we utilize an ensemble of projections from the Coupled Model Intercomparison Project Phase 5 (CMIP5) to develop probabilistic climate information, with and without an accounting of inter-model correlations, and use it to estimate climate-related risks to a local water utility in Colorado, U.S. We show that the tail risk of extreme climate changes in both mean precipitation and temperature is underestimated if model correlations are ignored. When coupled with impact models of the hydrology and infrastructure of the water utility, the underestimation of extreme climate changes substantially alters the quantification of risk for water supply shortages by mid-century. We argue that progress in climate change adaptation for local systems requires the recognition that there is less information in multi-model climate ensembles than previously thought. Importantly, adaptation decisions cannot be limited to the spread in one generation of climate models.

Probabilistic techniques for obtaining accurate patient counts in Clinical Data Warehouses

PubMed Central

Myers, Risa B.; Herskovic, Jorge R.

2011-01-01

Proposal and execution of clinical trials, computation of quality measures and discovery of correlation between medical phenomena are all applications where an accurate count of patients is needed. However, existing sources of this type of patient information, including Clinical Data Warehouses (CDW) may be incomplete or inaccurate. This research explores applying probabilistic techniques, supported by the MayBMS probabilistic database, to obtain accurate patient counts from a clinical data warehouse containing synthetic patient data. We present a synthetic clinical data warehouse (CDW), and populate it with simulated data using a custom patient data generation engine. We then implement, evaluate and compare different techniques for obtaining patients counts. We model billing as a test for the presence of a condition. We compute billing’s sensitivity and specificity both by conducting a “Simulated Expert Review” where a representative sample of records are reviewed and labeled by experts, and by obtaining the ground truth for every record. We compute the posterior probability of a patient having a condition through a “Bayesian Chain”, using Bayes’ Theorem to calculate the probability of a patient having a condition after each visit. The second method is a “one-shot” approach that computes the probability of a patient having a condition based on whether the patient is ever billed for the condition Our results demonstrate the utility of probabilistic approaches, which improve on the accuracy of raw counts. In particular, the simulated review paired with a single application of Bayes’ Theorem produces the best results, with an average error rate of 2.1% compared to 43.7% for the straightforward billing counts. Overall, this research demonstrates that Bayesian probabilistic approaches improve patient counts on simulated patient populations. We believe that total patient counts based on billing data are one of the many possible applications of our Bayesian framework. Use of these probabilistic techniques will enable more accurate patient counts and better results for applications requiring this metric. PMID:21986292

Statistical appearance models based on probabilistic correspondences.

PubMed

Krüger, Julia; Ehrhardt, Jan; Handels, Heinz

2017-04-01

Model-based image analysis is indispensable in medical image processing. One key aspect of building statistical shape and appearance models is the determination of one-to-one correspondences in the training data set. At the same time, the identification of these correspondences is the most challenging part of such methods. In our earlier work, we developed an alternative method using correspondence probabilities instead of exact one-to-one correspondences for a statistical shape model (Hufnagel et al., 2008). In this work, a new approach for statistical appearance models without one-to-one correspondences is proposed. A sparse image representation is used to build a model that combines point position and appearance information at the same time. Probabilistic correspondences between the derived multi-dimensional feature vectors are used to omit the need for extensive preprocessing of finding landmarks and correspondences as well as to reduce the dependence of the generated model on the landmark positions. Model generation and model fitting can now be expressed by optimizing a single global criterion derived from a maximum a-posteriori (MAP) approach with respect to model parameters that directly affect both shape and appearance of the considered objects inside the images. The proposed approach describes statistical appearance modeling in a concise and flexible mathematical framework. Besides eliminating the demand for costly correspondence determination, the method allows for additional constraints as topological regularity in the modeling process. In the evaluation the model was applied for segmentation and landmark identification in hand X-ray images. The results demonstrate the feasibility of the model to detect hand contours as well as the positions of the joints between finger bones for unseen test images. Further, we evaluated the model on brain data of stroke patients to show the ability of the proposed model to handle partially corrupted data and to demonstrate a possible employment of the correspondence probabilities to indicate these corrupted/pathological areas. Copyright © 2017 Elsevier B.V. All rights reserved.

Uncertainty estimation of long-range ensemble forecasts of snowmelt flood characteristics

NASA Astrophysics Data System (ADS)

Kuchment, L.

2012-04-01

Long-range forecasts of snowmelt flood characteristics with the lead time of 2-3 months have important significance for regulation of flood runoff and mitigation of flood damages at almost all large Russian rivers At the same time, the application of current forecasting techniques based on regression relationships between the runoff volume and the indexes of river basin conditions can lead to serious errors in forecasting resulted in large economic losses caused by wrong flood regulation. The forecast errors can be caused by complicated processes of soil freezing and soil moisture redistribution, too high rate of snow melt, large liquid precipitation before snow melt. or by large difference of meteorological conditions during the lead-time periods from climatologic ones. Analysis of economic losses had shown that the largest damages could, to a significant extent, be avoided if the decision makers had an opportunity to take into account predictive uncertainty and could use more cautious strategies in runoff regulation. Development of methodology of long-range ensemble forecasting of spring/summer floods which is based on distributed physically-based runoff generation models has created, in principle, a new basis for improving hydrological predictions as well as for estimating their uncertainty. This approach is illustrated by forecasting of the spring-summer floods at the Vyatka River and the Seim River basins. The application of the physically - based models of snowmelt runoff generation give a essential improving of statistical estimates of the deterministic forecasts of the flood volume in comparison with the forecasts obtained from the regression relationships. These models had been used also for the probabilistic forecasts assigning meteorological inputs during lead time periods from the available historical daily series, and from the series simulated by using a weather generator and the Monte Carlo procedure. The weather generator consists of the stochastic models of daily temperature and precipitation. The performance of the probabilistic forecasts were estimated by the ranked probability skill scores. The application of Monte Carlo simulations using weather generator has given better results then using the historical meteorological series.

Application of a stochastic snowmelt model for probabilistic decisionmaking

NASA Technical Reports Server (NTRS)

Mccuen, R. H.

1983-01-01

A stochastic form of the snowmelt runoff model that can be used for probabilistic decision-making was developed. The use of probabilistic streamflow predictions instead of single valued deterministic predictions leads to greater accuracy in decisions. While the accuracy of the output function is important in decisionmaking, it is also important to understand the relative importance of the coefficients. Therefore, a sensitivity analysis was made for each of the coefficients.

Vagueness as Probabilistic Linguistic Knowledge

NASA Astrophysics Data System (ADS)

Lassiter, Daniel

Consideration of the metalinguistic effects of utterances involving vague terms has led Barker [1] to treat vagueness using a modified Stalnakerian model of assertion. I present a sorites-like puzzle for factual beliefs in the standard Stalnakerian model [28] and show that it can be resolved by enriching the model to make use of probabilistic belief spaces. An analogous problem arises for metalinguistic information in Barker's model, and I suggest that a similar enrichment is needed here as well. The result is a probabilistic theory of linguistic representation that retains a classical metalanguage but avoids the undesirable divorce between meaning and use inherent in the epistemic theory [34]. I also show that the probabilistic approach provides a plausible account of the sorites paradox and higher-order vagueness and that it fares well empirically and conceptually in comparison to leading competitors.

Probalistic Models for Solar Particle Events

NASA Technical Reports Server (NTRS)

Adams, James H., Jr.; Xapsos, Michael

2009-01-01

Probabilistic Models of Solar Particle Events (SPEs) are used in space mission design studies to describe the radiation environment that can be expected at a specified confidence level. The task of the designer is then to choose a design that will operate in the model radiation environment. Probabilistic models have already been developed for solar proton events that describe the peak flux, event-integrated fluence and missionintegrated fluence. In addition a probabilistic model has been developed that describes the mission-integrated fluence for the Z>2 elemental spectra. This talk will focus on completing this suite of models by developing models for peak flux and event-integrated fluence elemental spectra for the Z>2 element

EFFECTS OF CORRELATED PROBABILISTIC EXPOSURE MODEL INPUTS ON SIMULATED RESULTS

EPA Science Inventory

In recent years, more probabilistic models have been developed to quantify aggregate human exposures to environmental pollutants. The impact of correlation among inputs in these models is an important issue, which has not been resolved. Obtaining correlated data and implementi...

Dynamic Models Applied to Landslides: Study Case Angangueo, MICHOACÁN, MÉXICO.

NASA Astrophysics Data System (ADS)

Torres Fernandez, L.; Hernández Madrigal, V. M., , Dr; Capra, L.; Domínguez Mota, F. J., , Dr

2017-12-01

Most existing models for landslide zonification are static type, do not consider the dynamic behavior of the trigger factor. This results in a limited representation of the actual zonation of slope instability, present a short-term validity, cańt be applied for the design of early warning systems, etc. Particularly in Mexico, these models are static because they do not consider triggering factor such as precipitation. In this work, we present a numerical evaluation to know the landslide susceptibility, based on probabilistic methods. Which are based on the generation of time series, which are generated from the meteorological stations, having limited information an interpolation is made to generate the simulation of the precipitation in the zone. The obtained information is integrated in PCRaster and in conjunction with the conditioning factors it is possible to generate a dynamic model. This model will be applied for landslide zoning in the municipality of Angangueo, characterized by frequent logging of debris and mud flow, translational and rotational landslides, detonated by atypical precipitations, such as those recorded in 2010. These caused economic losses and humans. With these models, it would be possible to generate probable scenarios that help the Angangueo's population to reduce the risks and to carry out actions of constant resilience activities.

Influence of weak hip abductor muscles on joint contact forces during normal walking: probabilistic modeling analysis.

PubMed

Valente, Giordano; Taddei, Fulvia; Jonkers, Ilse

2013-09-03

The weakness of hip abductor muscles is related to lower-limb joint osteoarthritis, and joint overloading may increase the risk for disease progression. The relationship between muscle strength, structural joint deterioration and joint loading makes the latter an important parameter in the study of onset and follow-up of the disease. Since the relationship between hip abductor weakness and joint loading still remains an open question, the purpose of this study was to adopt a probabilistic modeling approach to give insights into how the weakness of hip abductor muscles, in the extent to which normal gait could be unaltered, affects ipsilateral joint contact forces. A generic musculoskeletal model was scaled to each healthy subject included in the study, and the maximum force-generating capacity of each hip abductor muscle in the model was perturbed to evaluate how all physiologically possible configurations of hip abductor weakness affected the joint contact forces during walking. In general, the muscular system was able to compensate for abductor weakness. The reduced force-generating capacity of the abductor muscles affected joint contact forces to a mild extent, with 50th percentile mean differences up to 0.5 BW (maximum 1.7 BW). There were greater increases in the peak knee joint loads than in loads at the hip or ankle. Gluteus medius, particularly the anterior compartment, was the abductor muscle with the most influence on hip and knee loads. Further studies should assess if these increases in joint loading may affect initiation and progression of osteoarthritis. Copyright © 2013 Elsevier Ltd. All rights reserved.

A review of predictive coding algorithms.

PubMed

Spratling, M W

2017-03-01

Predictive coding is a leading theory of how the brain performs probabilistic inference. However, there are a number of distinct algorithms which are described by the term "predictive coding". This article provides a concise review of these different predictive coding algorithms, highlighting their similarities and differences. Five algorithms are covered: linear predictive coding which has a long and influential history in the signal processing literature; the first neuroscience-related application of predictive coding to explaining the function of the retina; and three versions of predictive coding that have been proposed to model cortical function. While all these algorithms aim to fit a generative model to sensory data, they differ in the type of generative model they employ, in the process used to optimise the fit between the model and sensory data, and in the way that they are related to neurobiology. Copyright © 2016 Elsevier Inc. All rights reserved.

Differential theory of learning for efficient neural network pattern recognition

NASA Astrophysics Data System (ADS)

Hampshire, John B., II; Vijaya Kumar, Bhagavatula

1993-09-01

We describe a new theory of differential learning by which a broad family of pattern classifiers (including many well-known neural network paradigms) can learn stochastic concepts efficiently. We describe the relationship between a classifier's ability to generate well to unseen test examples and the efficiency of the strategy by which it learns. We list a series of proofs that differential learning is efficient in its information and computational resource requirements, whereas traditional probabilistic learning strategies are not. The proofs are illustrated by a simple example that lends itself to closed-form analysis. We conclude with an optical character recognition task for which three different types of differentially generated classifiers generalize significantly better than their probabilistically generated counterparts.

Use of limited data to construct Bayesian networks for probabilistic risk assessment.

DOE Office of Scientific and Technical Information (OSTI.GOV)

Groth, Katrina M.; Swiler, Laura Painton

2013-03-01

Probabilistic Risk Assessment (PRA) is a fundamental part of safety/quality assurance for nuclear power and nuclear weapons. Traditional PRA very effectively models complex hardware system risks using binary probabilistic models. However, traditional PRA models are not flexible enough to accommodate non-binary soft-causal factors, such as digital instrumentation&control, passive components, aging, common cause failure, and human errors. Bayesian Networks offer the opportunity to incorporate these risks into the PRA framework. This report describes the results of an early career LDRD project titled %E2%80%9CUse of Limited Data to Construct Bayesian Networks for Probabilistic Risk Assessment%E2%80%9D. The goal of the work was tomore » establish the capability to develop Bayesian Networks from sparse data, and to demonstrate this capability by producing a data-informed Bayesian Network for use in Human Reliability Analysis (HRA) as part of nuclear power plant Probabilistic Risk Assessment (PRA). This report summarizes the research goal and major products of the research.« less

Probabilistic drug connectivity mapping

PubMed Central

2014-01-01

Background The aim of connectivity mapping is to match drugs using drug-treatment gene expression profiles from multiple cell lines. This can be viewed as an information retrieval task, with the goal of finding the most relevant profiles for a given query drug. We infer the relevance for retrieval by data-driven probabilistic modeling of the drug responses, resulting in probabilistic connectivity mapping, and further consider the available cell lines as different data sources. We use a special type of probabilistic model to separate what is shared and specific between the sources, in contrast to earlier connectivity mapping methods that have intentionally aggregated all available data, neglecting information about the differences between the cell lines. Results We show that the probabilistic multi-source connectivity mapping method is superior to alternatives in finding functionally and chemically similar drugs from the Connectivity Map data set. We also demonstrate that an extension of the method is capable of retrieving combinations of drugs that match different relevant parts of the query drug response profile. Conclusions The probabilistic modeling-based connectivity mapping method provides a promising alternative to earlier methods. Principled integration of data from different cell lines helps to identify relevant responses for specific drug repositioning applications. PMID:24742351

Probabilistic analysis of preload in the abutment screw of a dental implant complex.

PubMed

Guda, Teja; Ross, Thomas A; Lang, Lisa A; Millwater, Harry R

2008-09-01

Screw loosening is a problem for a percentage of implants. A probabilistic analysis to determine the cumulative probability distribution of the preload, the probability of obtaining an optimal preload, and the probabilistic sensitivities identifying important variables is lacking. The purpose of this study was to examine the inherent variability of material properties, surface interactions, and applied torque in an implant system to determine the probability of obtaining desired preload values and to identify the significant variables that affect the preload. Using software programs, an abutment screw was subjected to a tightening torque and the preload was determined from finite element (FE) analysis. The FE model was integrated with probabilistic analysis software. Two probabilistic analysis methods (advanced mean value and Monte Carlo sampling) were applied to determine the cumulative distribution function (CDF) of preload. The coefficient of friction, elastic moduli, Poisson's ratios, and applied torque were modeled as random variables and defined by probability distributions. Separate probability distributions were determined for the coefficient of friction in well-lubricated and dry environments. The probabilistic analyses were performed and the cumulative distribution of preload was determined for each environment. A distinct difference was seen between the preload probability distributions generated in a dry environment (normal distribution, mean (SD): 347 (61.9) N) compared to a well-lubricated environment (normal distribution, mean (SD): 616 (92.2) N). The probability of obtaining a preload value within the target range was approximately 54% for the well-lubricated environment and only 0.02% for the dry environment. The preload is predominately affected by the applied torque and coefficient of friction between the screw threads and implant bore at lower and middle values of the preload CDF, and by the applied torque and the elastic modulus of the abutment screw at high values of the preload CDF. Lubrication at the threaded surfaces between the abutment screw and implant bore affects the preload developed in the implant complex. For the well-lubricated surfaces, only approximately 50% of implants will have preload values within the generally accepted range. This probability can be improved by applying a higher torque than normally recommended or a more closely controlled torque than typically achieved. It is also suggested that materials with higher elastic moduli be used in the manufacture of the abutment screw to achieve a higher preload.

Remembrance of inferences past: Amortization in human hypothesis generation.

PubMed

Dasgupta, Ishita; Schulz, Eric; Goodman, Noah D; Gershman, Samuel J

2018-05-21

Bayesian models of cognition assume that people compute probability distributions over hypotheses. However, the required computations are frequently intractable or prohibitively expensive. Since people often encounter many closely related distributions, selective reuse of computations (amortized inference) is a computationally efficient use of the brain's limited resources. We present three experiments that provide evidence for amortization in human probabilistic reasoning. When sequentially answering two related queries about natural scenes, participants' responses to the second query systematically depend on the structure of the first query. This influence is sensitive to the content of the queries, only appearing when the queries are related. Using a cognitive load manipulation, we find evidence that people amortize summary statistics of previous inferences, rather than storing the entire distribution. These findings support the view that the brain trades off accuracy and computational cost, to make efficient use of its limited cognitive resources to approximate probabilistic inference. Copyright © 2018 Elsevier B.V. All rights reserved.

Probabilistic Approach to Conditional Probability of Release of Hazardous Materials from Railroad Tank Cars during Accidents

DOT National Transportation Integrated Search

2009-10-13

This paper describes a probabilistic approach to estimate the conditional probability of release of hazardous materials from railroad tank cars during train accidents. Monte Carlo methods are used in developing a probabilistic model to simulate head ...

Distributed collaborative probabilistic design for turbine blade-tip radial running clearance using support vector machine of regression

NASA Astrophysics Data System (ADS)

Fei, Cheng-Wei; Bai, Guang-Chen

2014-12-01

To improve the computational precision and efficiency of probabilistic design for mechanical dynamic assembly like the blade-tip radial running clearance (BTRRC) of gas turbine, a distribution collaborative probabilistic design method-based support vector machine of regression (SR)(called as DCSRM) is proposed by integrating distribution collaborative response surface method and support vector machine regression model. The mathematical model of DCSRM is established and the probabilistic design idea of DCSRM is introduced. The dynamic assembly probabilistic design of aeroengine high-pressure turbine (HPT) BTRRC is accomplished to verify the proposed DCSRM. The analysis results reveal that the optimal static blade-tip clearance of HPT is gained for designing BTRRC, and improving the performance and reliability of aeroengine. The comparison of methods shows that the DCSRM has high computational accuracy and high computational efficiency in BTRRC probabilistic analysis. The present research offers an effective way for the reliability design of mechanical dynamic assembly and enriches mechanical reliability theory and method.

Improving Global Forecast System of extreme precipitation events with regional statistical model: Application of quantile-based probabilistic forecasts

NASA Astrophysics Data System (ADS)

Shastri, Hiteshri; Ghosh, Subimal; Karmakar, Subhankar

2017-02-01

Forecasting of extreme precipitation events at a regional scale is of high importance due to their severe impacts on society. The impacts are stronger in urban regions due to high flood potential as well high population density leading to high vulnerability. Although significant scientific improvements took place in the global models for weather forecasting, they are still not adequate at a regional scale (e.g., for an urban region) with high false alarms and low detection. There has been a need to improve the weather forecast skill at a local scale with probabilistic outcome. Here we develop a methodology with quantile regression, where the reliably simulated variables from Global Forecast System are used as predictors and different quantiles of rainfall are generated corresponding to that set of predictors. We apply this method to a flood-prone coastal city of India, Mumbai, which has experienced severe floods in recent years. We find significant improvements in the forecast with high detection and skill scores. We apply the methodology to 10 ensemble members of Global Ensemble Forecast System and find a reduction in ensemble uncertainty of precipitation across realizations with respect to that of original precipitation forecasts. We validate our model for the monsoon season of 2006 and 2007, which are independent of the training/calibration data set used in the study. We find promising results and emphasize to implement such data-driven methods for a better probabilistic forecast at an urban scale primarily for an early flood warning.

Solving probability reasoning based on DNA strand displacement and probability modules.

PubMed

Zhang, Qiang; Wang, Xiaobiao; Wang, Xiaojun; Zhou, Changjun

2017-12-01

In computation biology, DNA strand displacement technology is used to simulate the computation process and has shown strong computing ability. Most researchers use it to solve logic problems, but it is only rarely used in probabilistic reasoning. To process probabilistic reasoning, a conditional probability derivation model and total probability model based on DNA strand displacement were established in this paper. The models were assessed through the game "read your mind." It has been shown to enable the application of probabilistic reasoning in genetic diagnosis. Copyright © 2017 Elsevier Ltd. All rights reserved.

Network approaches for expert decisions in sports.

PubMed

Glöckner, Andreas; Heinen, Thomas; Johnson, Joseph G; Raab, Markus

2012-04-01

This paper focuses on a model comparison to explain choices based on gaze behavior via simulation procedures. We tested two classes of models, a parallel constraint satisfaction (PCS) artificial neuronal network model and an accumulator model in a handball decision-making task from a lab experiment. Both models predict action in an option-generation task in which options can be chosen from the perspective of a playmaker in handball (i.e., passing to another player or shooting at the goal). Model simulations are based on a dataset of generated options together with gaze behavior measurements from 74 expert handball players for 22 pieces of video footage. We implemented both classes of models as deterministic vs. probabilistic models including and excluding fitted parameters. Results indicated that both classes of models can fit and predict participants' initially generated options based on gaze behavior data, and that overall, the classes of models performed about equally well. Early fixations were thereby particularly predictive for choices. We conclude that the analyses of complex environments via network approaches can be successfully applied to the field of experts' decision making in sports and provide perspectives for further theoretical developments. Copyright © 2011 Elsevier B.V. All rights reserved.

Efficient Probabilistic Diagnostics for Electrical Power Systems

NASA Technical Reports Server (NTRS)

Mengshoel, Ole J.; Chavira, Mark; Cascio, Keith; Poll, Scott; Darwiche, Adnan; Uckun, Serdar

2008-01-01

We consider in this work the probabilistic approach to model-based diagnosis when applied to electrical power systems (EPSs). Our probabilistic approach is formally well-founded, as it based on Bayesian networks and arithmetic circuits. We investigate the diagnostic task known as fault isolation, and pay special attention to meeting two of the main challenges . model development and real-time reasoning . often associated with real-world application of model-based diagnosis technologies. To address the challenge of model development, we develop a systematic approach to representing electrical power systems as Bayesian networks, supported by an easy-to-use speci.cation language. To address the real-time reasoning challenge, we compile Bayesian networks into arithmetic circuits. Arithmetic circuit evaluation supports real-time diagnosis by being predictable and fast. In essence, we introduce a high-level EPS speci.cation language from which Bayesian networks that can diagnose multiple simultaneous failures are auto-generated, and we illustrate the feasibility of using arithmetic circuits, compiled from Bayesian networks, for real-time diagnosis on real-world EPSs of interest to NASA. The experimental system is a real-world EPS, namely the Advanced Diagnostic and Prognostic Testbed (ADAPT) located at the NASA Ames Research Center. In experiments with the ADAPT Bayesian network, which currently contains 503 discrete nodes and 579 edges, we .nd high diagnostic accuracy in scenarios where one to three faults, both in components and sensors, were inserted. The time taken to compute the most probable explanation using arithmetic circuits has a small mean of 0.2625 milliseconds and standard deviation of 0.2028 milliseconds. In experiments with data from ADAPT we also show that arithmetic circuit evaluation substantially outperforms joint tree propagation and variable elimination, two alternative algorithms for diagnosis using Bayesian network inference.

Ensemble assimilation of ARGO temperature profile, sea surface temperature and Altimetric satellite data into an eddy permitting primitive equation model of the North Atlantic ocean

NASA Astrophysics Data System (ADS)

Yan, Yajing; Barth, Alexander; Beckers, Jean-Marie; Candille, Guillem; Brankart, Jean-Michel; Brasseur, Pierre

2015-04-01

Sea surface height, sea surface temperature and temperature profiles at depth collected between January and December 2005 are assimilated into a realistic eddy permitting primitive equation model of the North Atlantic Ocean using the Ensemble Kalman Filter. 60 ensemble members are generated by adding realistic noise to the forcing parameters related to the temperature. The ensemble is diagnosed and validated by comparison between the ensemble spread and the model/observation difference, as well as by rank histogram before the assimilation experiments. Incremental analysis update scheme is applied in order to reduce spurious oscillations due to the model state correction. The results of the assimilation are assessed according to both deterministic and probabilistic metrics with observations used in the assimilation experiments and independent observations, which goes further than most previous studies and constitutes one of the original points of this paper. Regarding the deterministic validation, the ensemble means, together with the ensemble spreads are compared to the observations in order to diagnose the ensemble distribution properties in a deterministic way. Regarding the probabilistic validation, the continuous ranked probability score (CRPS) is used to evaluate the ensemble forecast system according to reliability and resolution. The reliability is further decomposed into bias and dispersion by the reduced centred random variable (RCRV) score in order to investigate the reliability properties of the ensemble forecast system. The improvement of the assimilation is demonstrated using these validation metrics. Finally, the deterministic validation and the probabilistic validation are analysed jointly. The consistency and complementarity between both validations are highlighted. High reliable situations, in which the RMS error and the CRPS give the same information, are identified for the first time in this paper.

Verification of recursive probabilistic integration (RPI) method for fatigue life management using non-destructive inspections

NASA Astrophysics Data System (ADS)

Chen, Tzikang J.; Shiao, Michael

2016-04-01

This paper verified a generic and efficient assessment concept for probabilistic fatigue life management. The concept is developed based on an integration of damage tolerance methodology, simulations methods1, 2, and a probabilistic algorithm RPI (recursive probability integration)3-9 considering maintenance for damage tolerance and risk-based fatigue life management. RPI is an efficient semi-analytical probabilistic method for risk assessment subjected to various uncertainties such as the variability in material properties including crack growth rate, initial flaw size, repair quality, random process modeling of flight loads for failure analysis, and inspection reliability represented by probability of detection (POD). In addition, unlike traditional Monte Carlo simulations (MCS) which requires a rerun of MCS when maintenance plan is changed, RPI can repeatedly use a small set of baseline random crack growth histories excluding maintenance related parameters from a single MCS for various maintenance plans. In order to fully appreciate the RPI method, a verification procedure was performed. In this study, MC simulations in the orders of several hundred billions were conducted for various flight conditions, material properties, and inspection scheduling, POD and repair/replacement strategies. Since the MC simulations are time-consuming methods, the simulations were conducted parallelly on DoD High Performance Computers (HPC) using a specialized random number generator for parallel computing. The study has shown that RPI method is several orders of magnitude more efficient than traditional Monte Carlo simulations.

Effect of time dependence on probabilistic seismic-hazard maps and deaggregation for the central Apennines, Italy

USGS Publications Warehouse

Akinci, A.; Galadini, F.; Pantosti, D.; Petersen, M.; Malagnini, L.; Perkins, D.

2009-01-01

We produce probabilistic seismic-hazard assessments for the central Apennines, Italy, using time-dependent models that are characterized using a Brownian passage time recurrence model. Using aperiodicity parameters, ?? of 0.3, 0.5, and 0.7, we examine the sensitivity of the probabilistic ground motion and its deaggregation to these parameters. For the seismic source model we incorporate both smoothed historical seismicity over the area and geological information on faults. We use the maximum magnitude model for the fault sources together with a uniform probability of rupture along the fault (floating fault model) to model fictitious faults to account for earthquakes that cannot be correlated with known geologic structural segmentation.

Error Discounting in Probabilistic Category Learning

PubMed Central

Craig, Stewart; Lewandowsky, Stephan; Little, Daniel R.

2011-01-01

Some current theories of probabilistic categorization assume that people gradually attenuate their learning in response to unavoidable error. However, existing evidence for this error discounting is sparse and open to alternative interpretations. We report two probabilistic-categorization experiments that investigated error discounting by shifting feedback probabilities to new values after different amounts of training. In both experiments, responding gradually became less responsive to errors, and learning was slowed for some time after the feedback shift. Both results are indicative of error discounting. Quantitative modeling of the data revealed that adding a mechanism for error discounting significantly improved the fits of an exemplar-based and a rule-based associative learning model, as well as of a recency-based model of categorization. We conclude that error discounting is an important component of probabilistic learning. PMID:21355666

Probabilistic interpretation of Peelle's pertinent puzzle and its resolution

DOE Office of Scientific and Technical Information (OSTI.GOV)

Hanson, Kenneth M.; Kawano, T.; Talou, P.

2004-01-01

Peelle's Pertinent Puzzle (PPP) states a seemingly plausible set of measurements with their covariance matrix, which produce an implausible answer. To answer the PPP question, we describe a reasonable experimental situation that is consistent with the PPP solution. The confusion surrounding the PPP arises in part because of its imprecise statement, which permits to a variety of interpretations and resulting answers, some of which seem implausible. We emphasize the importance of basing the analysis on an unambiguous probabilistic model that reflects the experimental situation. We present several different models of how the measurements quoted in the PPP problem could bemore » obtained, and interpret their solution in terms of a detailed probabilistic analysis. We suggest a probabilistic approach to handling uncertainties about which model to use.« less

Probabilistic Interpretation of Peelle's Pertinent Puzzle and its Resolution

DOE Office of Scientific and Technical Information (OSTI.GOV)

Hanson, Kenneth M.; Kawano, Toshihiko; Talou, Patrick

2005-05-24

Peelle's Pertinent Puzzle (PPP) states a seemingly plausible set of measurements with their covariance matrix, which produce an implausible answer. To answer the PPP question, we describe a reasonable experimental situation that is consistent with the PPP solution. The confusion surrounding the PPP arises in part because of its imprecise statement, which permits to a variety of interpretations and resulting answers, some of which seem implausible. We emphasize the importance of basing the analysis on an unambiguous probabilistic model that reflects the experimental situation. We present several different models of how the measurements quoted in the PPP problem could bemore » obtained, and interpret their solution in terms of a detailed probabilistic analysis. We suggest a probabilistic approach to handling uncertainties about which model to use.« less

Failed rib region prediction in a human body model during crash events with precrash braking.

PubMed

Guleyupoglu, B; Koya, B; Barnard, R; Gayzik, F S

2018-02-28

The objective of this study is 2-fold. We used a validated human body finite element model to study the predicted chest injury (focusing on rib fracture as a function of element strain) based on varying levels of simulated precrash braking. Furthermore, we compare deterministic and probabilistic methods of rib injury prediction in the computational model. The Global Human Body Models Consortium (GHBMC) M50-O model was gravity settled in the driver position of a generic interior equipped with an advanced 3-point belt and airbag. Twelve cases were investigated with permutations for failure, precrash braking system, and crash severity. The severities used were median (17 kph), severe (34 kph), and New Car Assessment Program (NCAP; 56.4 kph). Cases with failure enabled removed rib cortical bone elements once 1.8% effective plastic strain was exceeded. Alternatively, a probabilistic framework found in the literature was used to predict rib failure. Both the probabilistic and deterministic methods take into consideration location (anterior, lateral, and posterior). The deterministic method is based on a rubric that defines failed rib regions dependent on a threshold for contiguous failed elements. The probabilistic method depends on age-based strain and failure functions. Kinematics between both methods were similar (peak max deviation: ΔX head = 17 mm; ΔZ head = 4 mm; ΔX thorax = 5 mm; ΔZ thorax = 1 mm). Seat belt forces at the time of probabilistic failed region initiation were lower than those at deterministic failed region initiation. The probabilistic method for rib fracture predicted more failed regions in the rib (an analog for fracture) than the deterministic method in all but 1 case where they were equal. The failed region patterns between models are similar; however, there are differences that arise due to stress reduced from element elimination that cause probabilistic failed regions to continue to rise after no deterministic failed region would be predicted. Both the probabilistic and deterministic methods indicate similar trends with regards to the effect of precrash braking; however, there are tradeoffs. The deterministic failed region method is more spatially sensitive to failure and is more sensitive to belt loads. The probabilistic failed region method allows for increased capability in postprocessing with respect to age. The probabilistic failed region method predicted more failed regions than the deterministic failed region method due to force distribution differences.

The analysis of the possibility of using 10-minute rainfall series to determine the maximum rainfall amount with 5 minutes duration

NASA Astrophysics Data System (ADS)

Kaźmierczak, Bartosz; Wartalska, Katarzyna; Wdowikowski, Marcin; Kotowski, Andrzej

2017-11-01

Modern scientific research in the area of heavy rainfall analysis regarding to the sewerage design indicates the need to develop and use probabilistic rain models. One of the issues that remains to be resolved is the length of the shortest amount of rain to be analyzed. It is commonly believed that the best time is 5 minutes, while the least rain duration measured by the national services is often 10 or even 15 minutes. Main aim of this paper is to present the difference between probabilistic rainfall models results given from rainfall time series including and excluding 5 minutes rainfall duration. Analysis were made for long-time period from 1961-2010 on polish meteorological station Legnica. To develop best fitted to measurement rainfall data probabilistic model 4 probabilistic distributions were used. Results clearly indicates that models including 5 minutes rainfall duration remains more appropriate to use.

Probabilistic sensitivity analysis incorporating the bootstrap: an example comparing treatments for the eradication of Helicobacter pylori.

PubMed

Pasta, D J; Taylor, J L; Henning, J M

1999-01-01

Decision-analytic models are frequently used to evaluate the relative costs and benefits of alternative therapeutic strategies for health care. Various types of sensitivity analysis are used to evaluate the uncertainty inherent in the models. Although probabilistic sensitivity analysis is more difficult theoretically and computationally, the results can be much more powerful and useful than deterministic sensitivity analysis. The authors show how a Monte Carlo simulation can be implemented using standard software to perform a probabilistic sensitivity analysis incorporating the bootstrap. The method is applied to a decision-analytic model evaluating the cost-effectiveness of Helicobacter pylori eradication. The necessary steps are straightforward and are described in detail. The use of the bootstrap avoids certain difficulties encountered with theoretical distributions. The probabilistic sensitivity analysis provided insights into the decision-analytic model beyond the traditional base-case and deterministic sensitivity analyses and should become the standard method for assessing sensitivity.

INFERENCE BUILDING BLOCKS

DTIC Science & Technology

2018-02-15

address the problem that probabilistic inference algorithms are diÿcult and tedious to implement, by expressing them in terms of a small number of...building blocks, which are automatic transformations on probabilistic programs. On one hand, our curation of these building blocks reflects the way human...reasoning with low-level computational optimization, so the speed and accuracy of the generated solvers are competitive with state-of-the-art systems. 15

A Re-Unification of Two Competing Models for Document Retrieval.

ERIC Educational Resources Information Center

Bodoff, David

1999-01-01

Examines query-oriented versus document-oriented information retrieval and feedback learning. Highlights include a reunification of the two approaches for probabilistic document retrieval and for vector space model (VSM) retrieval; learning in VSM and in probabilistic models; multi-dimensional scaling; and ongoing field studies. (LRW)

Multivariate postprocessing techniques for probabilistic hydrological forecasting

NASA Astrophysics Data System (ADS)

Hemri, Stephan; Lisniak, Dmytro; Klein, Bastian

2016-04-01

Hydrologic ensemble forecasts driven by atmospheric ensemble prediction systems need statistical postprocessing in order to account for systematic errors in terms of both mean and spread. Runoff is an inherently multivariate process with typical events lasting from hours in case of floods to weeks or even months in case of droughts. This calls for multivariate postprocessing techniques that yield well calibrated forecasts in univariate terms and ensure a realistic temporal dependence structure at the same time. To this end, the univariate ensemble model output statistics (EMOS; Gneiting et al., 2005) postprocessing method is combined with two different copula approaches that ensure multivariate calibration throughout the entire forecast horizon. These approaches comprise ensemble copula coupling (ECC; Schefzik et al., 2013), which preserves the dependence structure of the raw ensemble, and a Gaussian copula approach (GCA; Pinson and Girard, 2012), which estimates the temporal correlations from training observations. Both methods are tested in a case study covering three subcatchments of the river Rhine that represent different sizes and hydrological regimes: the Upper Rhine up to the gauge Maxau, the river Moselle up to the gauge Trier, and the river Lahn up to the gauge Kalkofen. The results indicate that both ECC and GCA are suitable for modelling the temporal dependences of probabilistic hydrologic forecasts (Hemri et al., 2015). References Gneiting, T., A. E. Raftery, A. H. Westveld, and T. Goldman (2005), Calibrated probabilistic forecasting using ensemble model output statistics and minimum CRPS estimation, Monthly Weather Review, 133(5), 1098-1118, DOI: 10.1175/MWR2904.1. Hemri, S., D. Lisniak, and B. Klein, Multivariate postprocessing techniques for probabilistic hydrological forecasting, Water Resources Research, 51(9), 7436-7451, DOI: 10.1002/2014WR016473. Pinson, P., and R. Girard (2012), Evaluating the quality of scenarios of short-term wind power generation, Applied Energy, 96, 12-20, DOI: 10.1016/j.apenergy.2011.11.004. Schefzik, R., T. L. Thorarinsdottir, and T. Gneiting (2013), Uncertainty quantification in complex simulation models using ensemble copula coupling, Statistical Science, 28, 616-640, DOI: 10.1214/13-STS443.

A climate-based multivariate extreme emulator of met-ocean-hydrological events for coastal flooding

NASA Astrophysics Data System (ADS)

Camus, Paula; Rueda, Ana; Mendez, Fernando J.; Tomas, Antonio; Del Jesus, Manuel; Losada, Iñigo J.

2015-04-01

Atmosphere-ocean general circulation models (AOGCMs) are useful to analyze large-scale climate variability (long-term historical periods, future climate projections). However, applications such as coastal flood modeling require climate information at finer scale. Besides, flooding events depend on multiple climate conditions: waves, surge levels from the open-ocean and river discharge caused by precipitation. Therefore, a multivariate statistical downscaling approach is adopted to reproduce relationships between variables and due to its low computational cost. The proposed method can be considered as a hybrid approach which combines a probabilistic weather type downscaling model with a stochastic weather generator component. Predictand distributions are reproduced modeling the relationship with AOGCM predictors based on a physical division in weather types (Camus et al., 2012). The multivariate dependence structure of the predictand (extreme events) is introduced linking the independent marginal distributions of the variables by a probabilistic copula regression (Ben Ayala et al., 2014). This hybrid approach is applied for the downscaling of AOGCM data to daily precipitation and maximum significant wave height and storm-surge in different locations along the Spanish coast. Reanalysis data is used to assess the proposed method. A commonly predictor for the three variables involved is classified using a regression-guided clustering algorithm. The most appropriate statistical model (general extreme value distribution, pareto distribution) for daily conditions is fitted. Stochastic simulation of the present climate is performed obtaining the set of hydraulic boundary conditions needed for high resolution coastal flood modeling. References: Camus, P., Menéndez, M., Méndez, F.J., Izaguirre, C., Espejo, A., Cánovas, V., Pérez, J., Rueda, A., Losada, I.J., Medina, R. (2014b). A weather-type statistical downscaling framework for ocean wave climate. Journal of Geophysical Research, doi: 10.1002/2014JC010141. Ben Ayala, M.A., Chebana, F., Ouarda, T.B.M.J. (2014). Probabilistic Gaussian Copula Regression Model for Multisite and Multivariable Downscaling, Journal of Climate, 27, 3331-3347.

Evaluating the Performance of a Climate-Driven Mortality Model during Heat Waves and Cold Spells in Europe

PubMed Central

Lowe, Rachel; Ballester, Joan; Creswick, James; Robine, Jean-Marie; Herrmann, François R.; Rodó, Xavier

2015-01-01

The impact of climate change on human health is a serious concern. In particular, changes in the frequency and intensity of heat waves and cold spells are of high relevance in terms of mortality and morbidity. This demonstrates the urgent need for reliable early-warning systems to help authorities prepare and respond to emergency situations. In this study, we evaluate the performance of a climate-driven mortality model to provide probabilistic predictions of exceeding emergency mortality thresholds for heat wave and cold spell scenarios. Daily mortality data corresponding to 187 NUTS2 regions across 16 countries in Europe were obtained from 1998–2003. Data were aggregated to 54 larger regions in Europe, defined according to similarities in population structure and climate. Location-specific average mortality rates, at given temperature intervals over the time period, were modelled to account for the increased mortality observed during both high and low temperature extremes and differing comfort temperatures between regions. Model parameters were estimated in a Bayesian framework, in order to generate probabilistic simulations of mortality across Europe for time periods of interest. For the heat wave scenario (1–15 August 2003), the model was successfully able to anticipate the occurrence or non-occurrence of mortality rates exceeding the emergency threshold (75th percentile of the mortality distribution) for 89% of the 54 regions, given a probability decision threshold of 70%. For the cold spell scenario (1–15 January 2003), mortality events in 69% of the regions were correctly anticipated with a probability decision threshold of 70%. By using a more conservative decision threshold of 30%, this proportion increased to 87%. Overall, the model performed better for the heat wave scenario. By replacing observed temperature data in the model with forecast temperature, from state-of-the-art European forecasting systems, probabilistic mortality predictions could potentially be made several months ahead of imminent heat waves and cold spells. PMID:25625407

Evaluating the performance of a climate-driven mortality model during heat waves and cold spells in Europe.

PubMed

Lowe, Rachel; Ballester, Joan; Creswick, James; Robine, Jean-Marie; Herrmann, François R; Rodó, Xavier

2015-01-23

The impact of climate change on human health is a serious concern. In particular, changes in the frequency and intensity of heat waves and cold spells are of high relevance in terms of mortality and morbidity. This demonstrates the urgent need for reliable early-warning systems to help authorities prepare and respond to emergency situations. In this study, we evaluate the performance of a climate-driven mortality model to provide probabilistic predictions of exceeding emergency mortality thresholds for heat wave and cold spell scenarios. Daily mortality data corresponding to 187 NUTS2 regions across 16 countries in Europe were obtained from 1998-2003. Data were aggregated to 54 larger regions in Europe, defined according to similarities in population structure and climate. Location-specific average mortality rates, at given temperature intervals over the time period, were modelled to account for the increased mortality observed during both high and low temperature extremes and differing comfort temperatures between regions. Model parameters were estimated in a Bayesian framework, in order to generate probabilistic simulations of mortality across Europe for time periods of interest. For the heat wave scenario (1-15 August 2003), the model was successfully able to anticipate the occurrence or non-occurrence of mortality rates exceeding the emergency threshold (75th percentile of the mortality distribution) for 89% of the 54 regions, given a probability decision threshold of 70%. For the cold spell scenario (1-15 January 2003), mortality events in 69% of the regions were correctly anticipated with a probability decision threshold of 70%. By using a more conservative decision threshold of 30%, this proportion increased to 87%. Overall, the model performed better for the heat wave scenario. By replacing observed temperature data in the model with forecast temperature, from state-of-the-art European forecasting systems, probabilistic mortality predictions could potentially be made several months ahead of imminent heat waves and cold spells.

The probabilistic nature of preferential choice.

PubMed

Rieskamp, Jörg

2008-11-01

Previous research has developed a variety of theories explaining when and why people's decisions under risk deviate from the standard economic view of expected utility maximization. These theories are limited in their predictive accuracy in that they do not explain the probabilistic nature of preferential choice, that is, why an individual makes different choices in nearly identical situations, or why the magnitude of these inconsistencies varies in different situations. To illustrate the advantage of probabilistic theories, three probabilistic theories of decision making under risk are compared with their deterministic counterparts. The probabilistic theories are (a) a probabilistic version of a simple choice heuristic, (b) a probabilistic version of cumulative prospect theory, and (c) decision field theory. By testing the theories with the data from three experimental studies, the superiority of the probabilistic models over their deterministic counterparts in predicting people's decisions under risk become evident. When testing the probabilistic theories against each other, decision field theory provides the best account of the observed behavior.

Probabilistic Prediction of Lifetimes of Ceramic Parts

NASA Technical Reports Server (NTRS)

Nemeth, Noel N.; Gyekenyesi, John P.; Jadaan, Osama M.; Palfi, Tamas; Powers, Lynn; Reh, Stefan; Baker, Eric H.

2006-01-01

ANSYS/CARES/PDS is a software system that combines the ANSYS Probabilistic Design System (PDS) software with a modified version of the Ceramics Analysis and Reliability Evaluation of Structures Life (CARES/Life) Version 6.0 software. [A prior version of CARES/Life was reported in Program for Evaluation of Reliability of Ceramic Parts (LEW-16018), NASA Tech Briefs, Vol. 20, No. 3 (March 1996), page 28.] CARES/Life models effects of stochastic strength, slow crack growth, and stress distribution on the overall reliability of a ceramic component. The essence of the enhancement in CARES/Life 6.0 is the capability to predict the probability of failure using results from transient finite-element analysis. ANSYS PDS models the effects of uncertainty in material properties, dimensions, and loading on the stress distribution and deformation. ANSYS/CARES/PDS accounts for the effects of probabilistic strength, probabilistic loads, probabilistic material properties, and probabilistic tolerances on the lifetime and reliability of the component. Even failure probability becomes a stochastic quantity that can be tracked as a response variable. ANSYS/CARES/PDS enables tracking of all stochastic quantities in the design space, thereby enabling more precise probabilistic prediction of lifetimes of ceramic components.

State Tracking and Fault Diagnosis for Dynamic Systems Using Labeled Uncertainty Graph.

PubMed

Zhou, Gan; Feng, Wenquan; Zhao, Qi; Zhao, Hongbo

2015-11-05

Cyber-physical systems such as autonomous spacecraft, power plants and automotive systems become more vulnerable to unanticipated failures as their complexity increases. Accurate tracking of system dynamics and fault diagnosis are essential. This paper presents an efficient state estimation method for dynamic systems modeled as concurrent probabilistic automata. First, the Labeled Uncertainty Graph (LUG) method in the planning domain is introduced to describe the state tracking and fault diagnosis processes. Because the system model is probabilistic, the Monte Carlo technique is employed to sample the probability distribution of belief states. In addition, to address the sample impoverishment problem, an innovative look-ahead technique is proposed to recursively generate most likely belief states without exhaustively checking all possible successor modes. The overall algorithms incorporate two major steps: a roll-forward process that estimates system state and identifies faults, and a roll-backward process that analyzes possible system trajectories once the faults have been detected. We demonstrate the effectiveness of this approach by applying it to a real world domain: the power supply control unit of a spacecraft.

Probabilistic Multi-Person Tracking Using Dynamic Bayes Networks

NASA Astrophysics Data System (ADS)

Klinger, T.; Rottensteiner, F.; Heipke, C.

2015-08-01

Tracking-by-detection is a widely used practice in recent tracking systems. These usually rely on independent single frame detections that are handled as observations in a recursive estimation framework. If these observations are imprecise the generated trajectory is prone to be updated towards a wrong position. In contrary to existing methods our novel approach uses a Dynamic Bayes Network in which the state vector of a recursive Bayes filter, as well as the location of the tracked object in the image are modelled as unknowns. These unknowns are estimated in a probabilistic framework taking into account a dynamic model, and a state-of-the-art pedestrian detector and classifier. The classifier is based on the Random Forest-algorithm and is capable of being trained incrementally so that new training samples can be incorporated at runtime. This allows the classifier to adapt to the changing appearance of a target and to unlearn outdated features. The approach is evaluated on a publicly available benchmark. The results confirm that our approach is well suited for tracking pedestrians over long distances while at the same time achieving comparatively good geometric accuracy.

Probabilistic atlas and geometric variability estimation to drive tissue segmentation.

PubMed

Xu, Hao; Thirion, Bertrand; Allassonnière, Stéphanie

2014-09-10

Computerized anatomical atlases play an important role in medical image analysis. While an atlas usually refers to a standard or mean image also called template, which presumably represents well a given population, it is not enough to characterize the observed population in detail. A template image should be learned jointly with the geometric variability of the shapes represented in the observations. These two quantities will in the sequel form the atlas of the corresponding population. The geometric variability is modeled as deformations of the template image so that it fits the observations. In this paper, we provide a detailed analysis of a new generative statistical model based on dense deformable templates that represents several tissue types observed in medical images. Our atlas contains both an estimation of probability maps of each tissue (called class) and the deformation metric. We use a stochastic algorithm for the estimation of the probabilistic atlas given a dataset. This atlas is then used for atlas-based segmentation method to segment the new images. Experiments are shown on brain T1 MRI datasets. Copyright © 2014 John Wiley & Sons, Ltd.

Probabilistic structural analysis of aerospace components using NESSUS

NASA Technical Reports Server (NTRS)

Shiao, Michael C.; Nagpal, Vinod K.; Chamis, Christos C.

1988-01-01

Probabilistic structural analysis of a Space Shuttle main engine turbopump blade is conducted using the computer code NESSUS (numerical evaluation of stochastic structures under stress). The goal of the analysis is to derive probabilistic characteristics of blade response given probabilistic descriptions of uncertainties in blade geometry, material properties, and temperature and pressure distributions. Probability densities are derived for critical blade responses. Risk assessment and failure life analysis is conducted assuming different failure models.

Cognitive Development Effects of Teaching Probabilistic Decision Making to Middle School Students

ERIC Educational Resources Information Center

Mjelde, James W.; Litzenberg, Kerry K.; Lindner, James R.

2011-01-01

This study investigated the comprehension and effectiveness of teaching formal, probabilistic decision-making skills to middle school students. Two specific objectives were to determine (1) if middle school students can comprehend a probabilistic decision-making approach, and (2) if exposure to the modeling approaches improves middle school…

Assessment of Coastal Communities' Vulnerability to Hurricane Surge under Climate Change via Probabilistic Map - A Case Study of the Southwest Coast of Florida

NASA Astrophysics Data System (ADS)

Feng, X.; Shen, S.

2014-12-01

The US coastline, over the past few years, has been overwhelmed by major storms including Hurricane Katrina (2005), Ike (2008), Irene (2011), and Sandy (2012). Supported by a growing and extensive body of evidence, a majority of research agrees hurricane activities have been enhanced due to climate change. However, the precise prediction of hurricane induced inundation remains a challenge. This study proposed a probabilistic inundation map based on a Statistically Modeled Storm Database (SMSD) to assess the probabilistic coastal inundation risk of Southwest Florida for near-future (20 years) scenario considering climate change. This map was processed through a Joint Probability Method with Optimal-Sampling (JPM-OS), developed by Condon and Sheng in 2012, and accompanied by a high resolution storm surge modeling system CH3D-SSMS. The probabilistic inundation map shows a 25.5-31.2% increase in spatially averaged inundation height compared to an inundation map of present-day scenario. To estimate climate change impacts on coastal communities, socioeconomic analyses were conducted using both the SMSD based probabilistic inundation map and the present-day inundation map. Combined with 2010 census data and 2012 parcel data from Florida Geographic Data Library, the differences of economic loss between the near-future and present day scenarios were used to generate an economic exposure map at census block group level to reflect coastal communities' exposure to climate change. The results show that climate change induced inundation increase has significant economic impacts. Moreover, the impacts are not equally distributed among different social groups considering their social vulnerability to hazards. Social vulnerability index at census block group level were obtained from Hazards and Vulnerability Research Institute. The demographic and economic variables in the index represent a community's adaptability to hazards. Local Moran's I was calculated to identify the clusters of highly exposed and vulnerable communities. The economic-exposure cluster map was overlapped with social-vulnerability cluster map to identify communities with low adaptive capability but high exposure. The result provides decision makers an intuitive tool to identify most susceptible communities for adaptation.

A probabilistic union model with automatic order selection for noisy speech recognition.

PubMed

Jancovic, P; Ming, J

2001-09-01

A critical issue in exploiting the potential of the sub-band-based approach to robust speech recognition is the method of combining the sub-band observations, for selecting the bands unaffected by noise. A new method for this purpose, i.e., the probabilistic union model, was recently introduced. This model has been shown to be capable of dealing with band-limited corruption, requiring no knowledge about the band position and statistical distribution of the noise. A parameter within the model, which we call its order, gives the best results when it equals the number of noisy bands. Since this information may not be available in practice, in this paper we introduce an automatic algorithm for selecting the order, based on the state duration pattern generated by the hidden Markov model (HMM). The algorithm has been tested on the TIDIGITS database corrupted by various types of additive band-limited noise with unknown noisy bands. The results have shown that the union model equipped with the new algorithm can achieve a recognition performance similar to that achieved when the number of noisy bands is known. The results show a very significant improvement over the traditional full-band model, without requiring prior information on either the position or the number of noisy bands. The principle of the algorithm for selecting the order based on state duration may also be applied to other sub-band combination methods.

The Terrestrial Investigation Model: A probabilistic risk assessment model for birds exposed to pesticides

EPA Science Inventory

One of the major recommendations of the National Academy of Science to the USEPA, NMFS and USFWS was to utilize probabilistic methods when assessing the risks of pesticides to federally listed endangered and threatened species. The Terrestrial Investigation Model (TIM, version 3....

Exploring Term Dependences in Probabilistic Information Retrieval Model.

ERIC Educational Resources Information Center

Cho, Bong-Hyun; Lee, Changki; Lee, Gary Geunbae

2003-01-01

Describes a theoretic process to apply Bahadur-Lazarsfeld expansion (BLE) to general probabilistic models and the state-of-the-art 2-Poisson model. Through experiments on two standard document collections, one in Korean and one in English, it is demonstrated that incorporation of term dependences using BLE significantly contributes to performance…

A Probabilistic Model of Phonological Relationships from Contrast to Allophony

ERIC Educational Resources Information Center

Hall, Kathleen Currie

2009-01-01

This dissertation proposes a model of phonological relationships, the Probabilistic Phonological Relationship Model (PPRM), that quantifies how predictably distributed two sounds in a relationship are. It builds on a core premise of traditional phonological analysis, that the ability to define phonological relationships such as contrast and…

Application of Generative Topographic Mapping to Gear Failures Monitoring

NASA Astrophysics Data System (ADS)

Liao, Guanglan; Li, Weihua; Shi, Tielin; Rao, Raj B. K. N.

2002-07-01

The Generative Topographic Mapping (GTM) model is introduced as a probabilistic re-formation of the self-organizing map and has already been used in a variety of applications. This paper presents a study of the GTM in industrial gear failures monitoring. Vibration signals are analyzed using the GTM model, and the results show that gear feature data sets can be projected into a two-dimensional space and clustered in different areas according to their conditions, which can classify and identify clearly a gear work condition with cracked or broken tooth compared with the normal condition. With the trace of the image points in the two-dimensional space, the variation of gear work conditions can be observed visually, therefore, the occurrence and varying trend of gear failures can be monitored in time.

Supervised Gamma Process Poisson Factorization

DOE Office of Scientific and Technical Information (OSTI.GOV)

Anderson, Dylan Zachary

This thesis develops the supervised gamma process Poisson factorization (S- GPPF) framework, a novel supervised topic model for joint modeling of count matrices and document labels. S-GPPF is fully generative and nonparametric: document labels and count matrices are modeled under a uni ed probabilistic framework and the number of latent topics is controlled automatically via a gamma process prior. The framework provides for multi-class classification of documents using a generative max-margin classifier. Several recent data augmentation techniques are leveraged to provide for exact inference using a Gibbs sampling scheme. The first portion of this thesis reviews supervised topic modeling andmore » several key mathematical devices used in the formulation of S-GPPF. The thesis then introduces the S-GPPF generative model and derives the conditional posterior distributions of the latent variables for posterior inference via Gibbs sampling. The S-GPPF is shown to exhibit state-of-the-art performance for joint topic modeling and document classification on a dataset of conference abstracts, beating out competing supervised topic models. The unique properties of S-GPPF along with its competitive performance make it a novel contribution to supervised topic modeling.« less

PubMed related articles: a probabilistic topic-based model for content similarity

PubMed Central

Lin, Jimmy; Wilbur, W John

2007-01-01

Background We present a probabilistic topic-based model for content similarity called pmra that underlies the related article search feature in PubMed. Whether or not a document is about a particular topic is computed from term frequencies, modeled as Poisson distributions. Unlike previous probabilistic retrieval models, we do not attempt to estimate relevance–but rather our focus is "relatedness", the probability that a user would want to examine a particular document given known interest in another. We also describe a novel technique for estimating parameters that does not require human relevance judgments; instead, the process is based on the existence of MeSH ® in MEDLINE ®. Results The pmra retrieval model was compared against bm25, a competitive probabilistic model that shares theoretical similarities. Experiments using the test collection from the TREC 2005 genomics track shows a small but statistically significant improvement of pmra over bm25 in terms of precision. Conclusion Our experiments suggest that the pmra model provides an effective ranking algorithm for related article search. PMID:17971238

Toward a Probabilistic Phenological Model for Wheat Growing Degree Days (GDD)

NASA Astrophysics Data System (ADS)

Rahmani, E.; Hense, A.

2017-12-01

Are there deterministic relations between phenological and climate parameters? The answer is surely `No'. This answer motivated us to solve the problem through probabilistic theories. Thus, we developed a probabilistic phenological model which has the advantage of giving additional information in terms of uncertainty. To that aim, we turned to a statistical analysis named survival analysis. Survival analysis deals with death in biological organisms and failure in mechanical systems. In survival analysis literature, death or failure is considered as an event. By event, in this research we mean ripening date of wheat. We will assume only one event in this special case. By time, we mean the growing duration from sowing to ripening as lifetime for wheat which is a function of GDD. To be more precise we will try to perform the probabilistic forecast for wheat ripening. The probability value will change between 0 and 1. Here, the survivor function gives the probability that the not ripened wheat survives longer than a specific time or will survive to the end of its lifetime as a ripened crop. The survival function at each station is determined by fitting a normal distribution to the GDD as the function of growth duration. Verification of the models obtained is done using CRPS skill score (CRPSS). The positive values of CRPSS indicate the large superiority of the probabilistic phonologic survival model to the deterministic models. These results demonstrate that considering uncertainties in modeling are beneficial, meaningful and necessary. We believe that probabilistic phenological models have the potential to help reduce the vulnerability of agricultural production systems to climate change thereby increasing food security.

Simulation optimization of PSA-threshold based prostate cancer screening policies

PubMed Central

Zhang, Jingyu; Denton, Brian T.; Shah, Nilay D.; Inman, Brant A.

2013-01-01

We describe a simulation optimization method to design PSA screening policies based on expected quality adjusted life years (QALYs). Our method integrates a simulation model in a genetic algorithm which uses a probabilistic method for selection of the best policy. We present computational results about the efficiency of our algorithm. The best policy generated by our algorithm is compared to previously recommended screening policies. Using the policies determined by our model, we present evidence that patients should be screened more aggressively but for a shorter length of time than previously published guidelines recommend. PMID:22302420

Towards Verification and Validation for Increased Autonomy

NASA Technical Reports Server (NTRS)

Giannakopoulou, Dimitra

2017-01-01

This presentation goes over the work we have performed over the last few years on verification and validation of the next generation onboard collision avoidance system, ACAS X, for commercial aircraft. It describes our work on probabilistic verification and synthesis of the model that ACAS X is based on, and goes on to the validation of that model with respect to actual simulation and flight data. The presentation then moves on to identify the characteristics of ACAS X that are related to autonomy and to discuss the challenges that autonomy pauses on VV. All work presented has already been published.

Learning a generative probabilistic grammar of experience: a process-level model of language acquisition.

PubMed

Kolodny, Oren; Lotem, Arnon; Edelman, Shimon

2015-03-01

We introduce a set of biologically and computationally motivated design choices for modeling the learning of language, or of other types of sequential, hierarchically structured experience and behavior, and describe an implemented system that conforms to these choices and is capable of unsupervised learning from raw natural-language corpora. Given a stream of linguistic input, our model incrementally learns a grammar that captures its statistical patterns, which can then be used to parse or generate new data. The grammar constructed in this manner takes the form of a directed weighted graph, whose nodes are recursively (hierarchically) defined patterns over the elements of the input stream. We evaluated the model in seventeen experiments, grouped into five studies, which examined, respectively, (a) the generative ability of grammar learned from a corpus of natural language, (b) the characteristics of the learned representation, (c) sequence segmentation and chunking, (d) artificial grammar learning, and (e) certain types of structure dependence. The model's performance largely vindicates our design choices, suggesting that progress in modeling language acquisition can be made on a broad front-ranging from issues of generativity to the replication of human experimental findings-by bringing biological and computational considerations, as well as lessons from prior efforts, to bear on the modeling approach. Copyright © 2014 Cognitive Science Society, Inc.

Global Infrasound Association Based on Probabilistic Clutter Categorization

NASA Astrophysics Data System (ADS)

Arora, Nimar; Mialle, Pierrick

2016-04-01

The IDC advances its methods and continuously improves its automatic system for the infrasound technology. The IDC focuses on enhancing the automatic system for the identification of valid signals and the optimization of the network detection threshold by identifying ways to refine signal characterization methodology and association criteria. An objective of this study is to reduce the number of associated infrasound arrivals that are rejected from the automatic bulletins when generating the reviewed event bulletins. Indeed, a considerable number of signal detections are due to local clutter sources such as microbaroms, waterfalls, dams, gas flares, surf (ocean breaking waves) etc. These sources are either too diffuse or too local to form events. Worse still, the repetitive nature of this clutter leads to a large number of false event hypotheses due to the random matching of clutter at multiple stations. Previous studies, for example [1], have worked on categorization of clutter using long term trends on detection azimuth, frequency, and amplitude at each station. In this work we continue the same line of reasoning to build a probabilistic model of clutter that is used as part of NETVISA [2], a Bayesian approach to network processing. The resulting model is a fusion of seismic, hydroacoustic and infrasound processing built on a unified probabilistic framework. References: [1] Infrasound categorization Towards a statistics based approach. J. Vergoz, P. Gaillard, A. Le Pichon, N. Brachet, and L. Ceranna. ITW 2011 [2] NETVISA: Network Processing Vertically Integrated Seismic Analysis. N. S. Arora, S. Russell, and E. Sudderth. BSSA 2013

Probabilistic Flood Maps to support decision-making: Mapping the Value of Information

NASA Astrophysics Data System (ADS)

Alfonso, L.; Mukolwe, M. M.; Di Baldassarre, G.

2016-02-01

Floods are one of the most frequent and disruptive natural hazards that affect man. Annually, significant flood damage is documented worldwide. Flood mapping is a common preimpact flood hazard mitigation measure, for which advanced methods and tools (such as flood inundation models) are used to estimate potential flood extent maps that are used in spatial planning. However, these tools are affected, largely to an unknown degree, by both epistemic and aleatory uncertainty. Over the past few years, advances in uncertainty analysis with respect to flood inundation modeling show that it is appropriate to adopt Probabilistic Flood Maps (PFM) to account for uncertainty. However, the following question arises; how can probabilistic flood hazard information be incorporated into spatial planning? Thus, a consistent framework to incorporate PFMs into the decision-making is required. In this paper, a novel methodology based on Decision-Making under Uncertainty theories, in particular Value of Information (VOI) is proposed. Specifically, the methodology entails the use of a PFM to generate a VOI map, which highlights floodplain locations where additional information is valuable with respect to available floodplain management actions and their potential consequences. The methodology is illustrated with a simplified example and also applied to a real case study in the South of France, where a VOI map is analyzed on the basis of historical land use change decisions over a period of 26 years. Results show that uncertain flood hazard information encapsulated in PFMs can aid decision-making in floodplain planning.

The Gain-Loss Model: A Probabilistic Skill Multimap Model for Assessing Learning Processes

ERIC Educational Resources Information Center

Robusto, Egidio; Stefanutti, Luca; Anselmi, Pasquale

2010-01-01

Within the theoretical framework of knowledge space theory, a probabilistic skill multimap model for assessing learning processes is proposed. The learning process of a student is modeled as a function of the student's knowledge and of an educational intervention on the attainment of specific skills required to solve problems in a knowledge…

A PROBABILISTIC MODELING FRAMEWORK FOR PREDICTING POPULATION EXPOSURES TO BENZENE

EPA Science Inventory

The US Environmental Protection Agency (EPA) is modifying their probabilistic Stochastic Human Exposure Dose Simulation (SHEDS) model to assess aggregate exposures to air toxics. Air toxics include urban Hazardous Air Pollutants (HAPS) such as benzene from mobile sources, part...

A new discriminative kernel from probabilistic models.

PubMed

Tsuda, Koji; Kawanabe, Motoaki; Rätsch, Gunnar; Sonnenburg, Sören; Müller, Klaus-Robert

2002-10-01

Recently, Jaakkola and Haussler (1999) proposed a method for constructing kernel functions from probabilistic models. Their so-called Fisher kernel has been combined with discriminative classifiers such as support vector machines and applied successfully in, for example, DNA and protein analysis. Whereas the Fisher kernel is calculated from the marginal log-likelihood, we propose the TOP kernel derived; from tangent vectors of posterior log-odds. Furthermore, we develop a theoretical framework on feature extractors from probabilistic models and use it for analyzing the TOP kernel. In experiments, our new discriminative TOP kernel compares favorably to the Fisher kernel.

Poisson-Like Spiking in Circuits with Probabilistic Synapses

PubMed Central

Moreno-Bote, Rubén

2014-01-01

Neuronal activity in cortex is variable both spontaneously and during stimulation, and it has the remarkable property that it is Poisson-like over broad ranges of firing rates covering from virtually zero to hundreds of spikes per second. The mechanisms underlying cortical-like spiking variability over such a broad continuum of rates are currently unknown. We show that neuronal networks endowed with probabilistic synaptic transmission, a well-documented source of variability in cortex, robustly generate Poisson-like variability over several orders of magnitude in their firing rate without fine-tuning of the network parameters. Other sources of variability, such as random synaptic delays or spike generation jittering, do not lead to Poisson-like variability at high rates because they cannot be sufficiently amplified by recurrent neuronal networks. We also show that probabilistic synapses predict Fano factor constancy of synaptic conductances. Our results suggest that synaptic noise is a robust and sufficient mechanism for the type of variability found in cortex. PMID:25032705

Improving inflow forecasting into hydropower reservoirs through a complementary modelling framework

NASA Astrophysics Data System (ADS)

Gragne, A. S.; Sharma, A.; Mehrotra, R.; Alfredsen, K.

2014-10-01

Accuracy of reservoir inflow forecasts is instrumental for maximizing the value of water resources and benefits gained through hydropower generation. Improving hourly reservoir inflow forecasts over a 24 h lead-time is considered within the day-ahead (Elspot) market of the Nordic exchange market. We present here a new approach for issuing hourly reservoir inflow forecasts that aims to improve on existing forecasting models that are in place operationally, without needing to modify the pre-existing approach, but instead formulating an additive or complementary model that is independent and captures the structure the existing model may be missing. Besides improving forecast skills of operational models, the approach estimates the uncertainty in the complementary model structure and produces probabilistic inflow forecasts that entrain suitable information for reducing uncertainty in the decision-making processes in hydropower systems operation. The procedure presented comprises an error model added on top of an un-alterable constant parameter conceptual model, the models being demonstrated with reference to the 207 km2 Krinsvatn catchment in central Norway. The structure of the error model is established based on attributes of the residual time series from the conceptual model. Deterministic and probabilistic evaluations revealed an overall significant improvement in forecast accuracy for lead-times up to 17 h. Season based evaluations indicated that the improvement in inflow forecasts varies across seasons and inflow forecasts in autumn and spring are less successful with the 95% prediction interval bracketing less than 95% of the observations for lead-times beyond 17 h.

A fluvial and pluvial probabilistic flood hazard analysis for Can Tho city, Vietnam

NASA Astrophysics Data System (ADS)

Apel, Heiko; Martinez, Oriol; Thi Chinh, Do; Viet Dung, Nguyen

2014-05-01

Can Tho city is the largest city and the economic heart of the Mekong Delta, Vietnam. Due to its economic importance and envisaged development goals the city grew rapidly in population size and extend over the last two decades. Large parts of the city are located in flood prone areas, and also the central parts of the city recently experienced an increasing number of flood events, both of fluvial and pluvial nature. As the economic power and asset values are constantly increasing, this poses a considerable risk for the city. The the aim of this study is to perform a flood hazard analysis considering both fluvial and pluvial floods and to derive probabilistic flood hazard maps. This requires in a first step an understanding of the typical flood mechanisms. Fluvial floods are triggered by a coincidence of high water levels during the annual flood period in the Mekong Delta with high tidal levels, which cause in combination short term inundations in Can Tho. Pluvial floods are triggered by typical tropical convective rain storms during the monsoon season. These two flood pathways are essentially independent in its sources and can thus be treated in the hazard analysis accordingly. For the fluvial hazard analysis we propose a bivariate frequency analysis of the Mekong flood characteristics, the annual maximum flood discharge Q and the annual flood volume V at the upper boundary of the Mekong Delta, the gauging station Kratie. This defines probabilities of exceedance of different Q-V pairs, which are transferred into synthetic flood hydrographs. The synthetic hydrographs are routed through a quasi-2D hydrodynamic model of the entire Mekong Delta in order to provide boundary conditions for a detailed hazard mapping of Can Tho. This downscaling step is necessary, because the huge complexity of the river and channel network does not allow for a proper definition of boundary conditions for Can Tho city by gauge data alone. In addition the available gauge data around Can Tho are too short for a meaningful frequency analysis. The detailed hazard mapping is performed by a 2D hydrodynamic model for Can Tho city. As the scenarios are derived in a Monte-Carlo framework, the final flood hazard maps are probabilistic, i.e. show the median flood hazard along with uncertainty estimates for each defined level of probabilities of exceedance. For the pluvial flood hazard a frequency analysis of the hourly rain gauge data of Can Tho is performed implementing a peak-over-threshold procedure. Based on this frequency analysis synthetic rains storms are generated in a Monte-Carlo framework for the same probabilities of exceedance as in the fluvial flood hazard analysis. Probabilistic flood hazard maps were then generated with the same 2D hydrodynamic model for the city. In a last step the fluvial and pluvial scenarios are combined assuming independence of the events. These scenarios were also transferred into hazard maps by the 2D hydrodynamic model finally yielding combined fluvial-pluvial probabilistic flood hazard maps for Can Tho. The derived set of maps may be used for an improved city planning or a flood risk analysis.

Probabilistic Usage of the Multi-Factor Interaction Model

NASA Technical Reports Server (NTRS)

Chamis, Christos C.

2008-01-01

A Multi-Factor Interaction Model (MFIM) is used to predict the insulating foam mass expulsion during the ascending of a space vehicle. The exponents in the MFIM are evaluated by an available approach which consists of least squares and an optimization algorithm. These results were subsequently used to probabilistically evaluate the effects of the uncertainties in each participating factor in the mass expulsion. The probabilistic results show that the surface temperature dominates at high probabilities and the pressure which causes the mass expulsion at low probabil

DOE Office of Scientific and Technical Information (OSTI.GOV)

Jimenez, O.; Departamento de Fisica, Facultad de Ciencias Basicas, Universidad de Antofagasta, Casilla 170, Antofagasta; Bergou, J.

We study the probabilistic cloning of three symmetric states. These states are defined by a single complex quantity, the inner product among them. We show that three different probabilistic cloning machines are necessary to optimally clone all possible families of three symmetric states. We also show that the optimal cloning probability of generating M copies out of one original can be cast as the quotient between the success probability of unambiguously discriminating one and M copies of symmetric states.

A probabilistic model for detecting rigid domains in protein structures.

PubMed

Nguyen, Thach; Habeck, Michael

2016-09-01

Large-scale conformational changes in proteins are implicated in many important biological functions. These structural transitions can often be rationalized in terms of relative movements of rigid domains. There is a need for objective and automated methods that identify rigid domains in sets of protein structures showing alternative conformational states. We present a probabilistic model for detecting rigid-body movements in protein structures. Our model aims to approximate alternative conformational states by a few structural parts that are rigidly transformed under the action of a rotation and a translation. By using Bayesian inference and Markov chain Monte Carlo sampling, we estimate all parameters of the model, including a segmentation of the protein into rigid domains, the structures of the domains themselves, and the rigid transformations that generate the observed structures. We find that our Gibbs sampling algorithm can also estimate the optimal number of rigid domains with high efficiency and accuracy. We assess the power of our method on several thousand entries of the DynDom database and discuss applications to various complex biomolecular systems. The Python source code for protein ensemble analysis is available at: https://github.com/thachnguyen/motion_detection : mhabeck@gwdg.de. © The Author 2016. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.

HIERARCHICAL PROBABILISTIC INFERENCE OF COSMIC SHEAR

DOE Office of Scientific and Technical Information (OSTI.GOV)

Schneider, Michael D.; Dawson, William A.; Hogg, David W.

2015-07-01

Point estimators for the shearing of galaxy images induced by gravitational lensing involve a complex inverse problem in the presence of noise, pixelization, and model uncertainties. We present a probabilistic forward modeling approach to gravitational lensing inference that has the potential to mitigate the biased inferences in most common point estimators and is practical for upcoming lensing surveys. The first part of our statistical framework requires specification of a likelihood function for the pixel data in an imaging survey given parameterized models for the galaxies in the images. We derive the lensing shear posterior by marginalizing over all intrinsic galaxymore » properties that contribute to the pixel data (i.e., not limited to galaxy ellipticities) and learn the distributions for the intrinsic galaxy properties via hierarchical inference with a suitably flexible conditional probabilitiy distribution specification. We use importance sampling to separate the modeling of small imaging areas from the global shear inference, thereby rendering our algorithm computationally tractable for large surveys. With simple numerical examples we demonstrate the improvements in accuracy from our importance sampling approach, as well as the significance of the conditional distribution specification for the intrinsic galaxy properties when the data are generated from an unknown number of distinct galaxy populations with different morphological characteristics.« less

A probabilistic approach for shallow rainfall-triggered landslide modeling at basin scale. A case study in the Luquillo Forest, Puerto Rico

NASA Astrophysics Data System (ADS)

Dialynas, Y. G.; Arnone, E.; Noto, L. V.; Bras, R. L.

2013-12-01

Slope stability depends on geotechnical and hydrological factors that exhibit wide natural spatial variability, yet sufficient measurements of the related parameters are rarely available over entire study areas. The uncertainty associated with the inability to fully characterize hydrologic behavior has an impact on any attempt to model landslide hazards. This work suggests a way to systematically account for this uncertainty in coupled distributed hydrological-stability models for shallow landslide hazard assessment. A probabilistic approach for the prediction of rainfall-triggered landslide occurrence at basin scale was implemented in an existing distributed eco-hydrological and landslide model, tRIBS-VEGGIE -landslide (Triangulated Irregular Network (TIN)-based Real-time Integrated Basin Simulator - VEGetation Generator for Interactive Evolution). More precisely, we upgraded tRIBS-VEGGIE- landslide to assess the likelihood of shallow landslides by accounting for uncertainty related to geotechnical and hydrological factors that directly affect slope stability. Natural variability of geotechnical soil characteristics was considered by randomizing soil cohesion and friction angle. Hydrological uncertainty related to the estimation of matric suction was taken into account by considering soil retention parameters as correlated random variables. The probability of failure is estimated through an assumed theoretical Factor of Safety (FS) distribution, conditioned on soil moisture content. At each cell, the temporally variant FS statistics are approximated by the First Order Second Moment (FOSM) method, as a function of parameters statistical properties. The model was applied on the Rio Mameyes Basin, located in the Luquillo Experimental Forest in Puerto Rico, where previous landslide analyses have been carried out. At each time step, model outputs include the probability of landslide occurrence across the basin, and the most probable depth of failure at each soil column. The use of the proposed probabilistic approach for shallow landslide prediction is able to reveal and quantify landslide risk at slopes assessed as stable by simpler deterministic methods.

Predicting Progression from Mild Cognitive Impairment to Alzheimer's Dementia Using Clinical, MRI, and Plasma Biomarkers via Probabilistic Pattern Classification

PubMed Central

Korolev, Igor O.; Symonds, Laura L.; Bozoki, Andrea C.

2016-01-01

Background Individuals with mild cognitive impairment (MCI) have a substantially increased risk of developing dementia due to Alzheimer's disease (AD). In this study, we developed a multivariate prognostic model for predicting MCI-to-dementia progression at the individual patient level. Methods Using baseline data from 259 MCI patients and a probabilistic, kernel-based pattern classification approach, we trained a classifier to distinguish between patients who progressed to AD-type dementia (n = 139) and those who did not (n = 120) during a three-year follow-up period. More than 750 variables across four data sources were considered as potential predictors of progression. These data sources included risk factors, cognitive and functional assessments, structural magnetic resonance imaging (MRI) data, and plasma proteomic data. Predictive utility was assessed using a rigorous cross-validation framework. Results Cognitive and functional markers were most predictive of progression, while plasma proteomic markers had limited predictive utility. The best performing model incorporated a combination of cognitive/functional markers and morphometric MRI measures and predicted progression with 80% accuracy (83% sensitivity, 76% specificity, AUC = 0.87). Predictors of progression included scores on the Alzheimer's Disease Assessment Scale, Rey Auditory Verbal Learning Test, and Functional Activities Questionnaire, as well as volume/cortical thickness of three brain regions (left hippocampus, middle temporal gyrus, and inferior parietal cortex). Calibration analysis revealed that the model is capable of generating probabilistic predictions that reliably reflect the actual risk of progression. Finally, we found that the predictive accuracy of the model varied with patient demographic, genetic, and clinical characteristics and could be further improved by taking into account the confidence of the predictions. Conclusions We developed an accurate prognostic model for predicting MCI-to-dementia progression over a three-year period. The model utilizes widely available, cost-effective, non-invasive markers and can be used to improve patient selection in clinical trials and identify high-risk MCI patients for early treatment. PMID:26901338

The impact of personalized probabilistic wall thickness models on peak wall stress in abdominal aortic aneurysms.

PubMed

Biehler, J; Wall, W A

2018-02-01

If computational models are ever to be used in high-stakes decision making in clinical practice, the use of personalized models and predictive simulation techniques is a must. This entails rigorous quantification of uncertainties as well as harnessing available patient-specific data to the greatest extent possible. Although researchers are beginning to realize that taking uncertainty in model input parameters into account is a necessity, the predominantly used probabilistic description for these uncertain parameters is based on elementary random variable models. In this work, we set out for a comparison of different probabilistic models for uncertain input parameters using the example of an uncertain wall thickness in finite element models of abdominal aortic aneurysms. We provide the first comparison between a random variable and a random field model for the aortic wall and investigate the impact on the probability distribution of the computed peak wall stress. Moreover, we show that the uncertainty about the prevailing peak wall stress can be reduced if noninvasively available, patient-specific data are harnessed for the construction of the probabilistic wall thickness model. Copyright © 2017 John Wiley & Sons, Ltd.

Fully probabilistic control for stochastic nonlinear control systems with input dependent noise.

PubMed

Herzallah, Randa

2015-03-01

Robust controllers for nonlinear stochastic systems with functional uncertainties can be consistently designed using probabilistic control methods. In this paper a generalised probabilistic controller design for the minimisation of the Kullback-Leibler divergence between the actual joint probability density function (pdf) of the closed loop control system, and an ideal joint pdf is presented emphasising how the uncertainty can be systematically incorporated in the absence of reliable systems models. To achieve this objective all probabilistic models of the system are estimated from process data using mixture density networks (MDNs) where all the parameters of the estimated pdfs are taken to be state and control input dependent. Based on this dependency of the density parameters on the input values, explicit formulations to the construction of optimal generalised probabilistic controllers are obtained through the techniques of dynamic programming and adaptive critic methods. Using the proposed generalised probabilistic controller, the conditional joint pdfs can be made to follow the ideal ones. A simulation example is used to demonstrate the implementation of the algorithm and encouraging results are obtained. Copyright © 2014 Elsevier Ltd. All rights reserved.

A Local Forecast of Land Surface Wetness Conditions, Drought, and St. Louis Encephalitis Virus Transmission Derived from Seasonal Climate Predictions

NASA Astrophysics Data System (ADS)

Shaman, J.; Stieglitz, M.; Zebiak, S.; Cane, M.; Day, J. F.

2002-12-01

We present an ensemble local hydrologic forecast derived from the seasonal forecasts of the International Research Institute (IRI) for Climate Prediction. Three- month seasonal forecasts were used to resample historical meteorological conditions and generate ensemble forcing datasets for a TOPMODEL-based hydrology model. Eleven retrospective forecasts were run at a Florida and New York site. Forecast skill was assessed for mean area modeled water table depth (WTD), i.e. near surface soil wetness conditions, and compared with WTD simulated with observed data. Hydrology model forecast skill was evident at the Florida site but not at the New York site. At the Florida site, persistence of hydrologic conditions and local skill of the IRI seasonal forecast contributed to the local hydrologic forecast skill. This forecast will permit probabilistic prediction of future hydrologic conditions. At the Florida site, we have also quantified the link between modeled WTD (i.e. drought) and the amplification and transmission of St. Louis Encephalitis virus (SLEV). We derive an empirical relationship between modeled land surface wetness and levels of SLEV transmission associated with human clinical cases. We then combine the seasonal forecasts of local, modeled WTD with this empirical relationship and produce retrospective probabilistic seasonal forecasts of epidemic SLEV transmission in Florida. Epidemic SLEV transmission forecast skill is demonstrated. These findings will permit real-time forecast of drought and resultant SLEV transmission in Florida.

Comparison of surface freshwater fluxes from different climate forecasts produced through different ensemble generation schemes.

NASA Astrophysics Data System (ADS)

Romanova, Vanya; Hense, Andreas; Wahl, Sabrina; Brune, Sebastian; Baehr, Johanna

2016-04-01

The decadal variability and its predictability of the surface net freshwater fluxes is compared in a set of retrospective predictions, all using the same model setup, and only differing in the implemented ocean initialisation method and ensemble generation method. The basic aim is to deduce the differences between the initialization/ensemble generation methods in view of the uncertainty of the verifying observational data sets. The analysis will give an approximation of the uncertainties of the net freshwater fluxes, which up to now appear to be one of the most uncertain products in observational data and model outputs. All ensemble generation methods are implemented into the MPI-ESM earth system model in the framework of the ongoing MiKlip project (www.fona-miklip.de). Hindcast experiments are initialised annually between 2000-2004, and from each start year 10 ensemble members are initialized for 5 years each. Four different ensemble generation methods are compared: (i) a method based on the Anomaly Transform method (Romanova and Hense, 2015) in which the initial oceanic perturbations represent orthogonal and balanced anomaly structures in space and time and between the variables taken from a control run, (ii) one-day-lagged ocean states from the MPI-ESM-LR baseline system (iii) one-day-lagged of ocean and atmospheric states with preceding full-field nudging to re-analysis in both the atmospheric and the oceanic component of the system - the baseline one MPI-ESM-LR system, (iv) an Ensemble Kalman Filter (EnKF) implemented into oceanic part of MPI-ESM (Brune et al. 2015), assimilating monthly subsurface oceanic temperature and salinity (EN3) using the Parallel Data Assimilation Framework (PDAF). The hindcasts are evaluated probabilistically using fresh water flux data sets from four different reanalysis data sets: MERRA, NCEP-R1, GFDL ocean reanalysis and GECCO2. The assessments show no clear differences in the evaluations scores on regional scales. However, on the global scale the physically motivated methods (i) and (iv) provide probabilistic hindcasts with a consistently higher reliability than the lagged initialization methods (ii)/(iii) despite the large uncertainties in the verifying observations and in the simulations.

GLISTRboost: Combining Multimodal MRI Segmentation, Registration, and Biophysical Tumor Growth Modeling with Gradient Boosting Machines for Glioma Segmentation.

PubMed

Bakas, Spyridon; Zeng, Ke; Sotiras, Aristeidis; Rathore, Saima; Akbari, Hamed; Gaonkar, Bilwaj; Rozycki, Martin; Pati, Sarthak; Davatzikos, Christos

2016-01-01

We present an approach for segmenting low- and high-grade gliomas in multimodal magnetic resonance imaging volumes. The proposed approach is based on a hybrid generative-discriminative model. Firstly, a generative approach based on an Expectation-Maximization framework that incorporates a glioma growth model is used to segment the brain scans into tumor, as well as healthy tissue labels. Secondly, a gradient boosting multi-class classification scheme is used to refine tumor labels based on information from multiple patients. Lastly, a probabilistic Bayesian strategy is employed to further refine and finalize the tumor segmentation based on patient-specific intensity statistics from the multiple modalities. We evaluated our approach in 186 cases during the training phase of the BRAin Tumor Segmentation (BRATS) 2015 challenge and report promising results. During the testing phase, the algorithm was additionally evaluated in 53 unseen cases, achieving the best performance among the competing methods.

Reconstructing Constructivism: Causal Models, Bayesian Learning Mechanisms, and the Theory Theory

ERIC Educational Resources Information Center

Gopnik, Alison; Wellman, Henry M.

2012-01-01

We propose a new version of the "theory theory" grounded in the computational framework of probabilistic causal models and Bayesian learning. Probabilistic models allow a constructivist but rigorous and detailed approach to cognitive development. They also explain the learning of both more specific causal hypotheses and more abstract framework…

QUANTIFYING AGGREGATE CHLORPYRIFOS EXPOSURE AND DOSE TO CHILDREN USING A PHYSICALLY-BASED TWO-STAGE MONTE CARLO PROBABILISTIC MODEL

EPA Science Inventory

To help address the Food Quality Protection Act of 1996, a physically-based, two-stage Monte Carlo probabilistic model has been developed to quantify and analyze aggregate exposure and dose to pesticides via multiple routes and pathways. To illustrate model capabilities and ide...

Probabilistic Analysis Techniques Applied to Complex Spacecraft Power System Modeling

NASA Technical Reports Server (NTRS)

Hojnicki, Jeffrey S.; Rusick, Jeffrey J.

2005-01-01

Electric power system performance predictions are critical to spacecraft, such as the International Space Station (ISS), to ensure that sufficient power is available to support all the spacecraft s power needs. In the case of the ISS power system, analyses to date have been deterministic, meaning that each analysis produces a single-valued result for power capability because of the complexity and large size of the model. As a result, the deterministic ISS analyses did not account for the sensitivity of the power capability to uncertainties in model input variables. Over the last 10 years, the NASA Glenn Research Center has developed advanced, computationally fast, probabilistic analysis techniques and successfully applied them to large (thousands of nodes) complex structural analysis models. These same techniques were recently applied to large, complex ISS power system models. This new application enables probabilistic power analyses that account for input uncertainties and produce results that include variations caused by these uncertainties. Specifically, N&R Engineering, under contract to NASA, integrated these advanced probabilistic techniques with Glenn s internationally recognized ISS power system model, System Power Analysis for Capability Evaluation (SPACE).

A Learning Framework for Winner-Take-All Networks with Stochastic Synapses.

PubMed

Mostafa, Hesham; Cauwenberghs, Gert

2018-06-01

Many recent generative models make use of neural networks to transform the probability distribution of a simple low-dimensional noise process into the complex distribution of the data. This raises the question of whether biological networks operate along similar principles to implement a probabilistic model of the environment through transformations of intrinsic noise processes. The intrinsic neural and synaptic noise processes in biological networks, however, are quite different from the noise processes used in current abstract generative networks. This, together with the discrete nature of spikes and local circuit interactions among the neurons, raises several difficulties when using recent generative modeling frameworks to train biologically motivated models. In this letter, we show that a biologically motivated model based on multilayer winner-take-all circuits and stochastic synapses admits an approximate analytical description. This allows us to use the proposed networks in a variational learning setting where stochastic backpropagation is used to optimize a lower bound on the data log likelihood, thereby learning a generative model of the data. We illustrate the generality of the proposed networks and learning technique by using them in a structured output prediction task and a semisupervised learning task. Our results extend the domain of application of modern stochastic network architectures to networks where synaptic transmission failure is the principal noise mechanism.

Solar Thermoelectricity via Advanced Latent Heat Storage: A Cost-Effective Small-Scale CSP Application

DOE Office of Scientific and Technical Information (OSTI.GOV)

Glatzmaier, Greg C.; Rea, J.; Olsen, Michele L.

We are developing a novel concentrating solar electricity-generating technology that is both modular and dispatchable. Solar ThermoElectricity via Advanced Latent heat Storage (STEALS) uses concentrated solar flux to generate high-temperature thermal energy, which directly converts to electricity via thermoelectric generators (TEGs), stored within a phase-change material (PCM) for electricity generation at a later time, or both allowing for simultaneous charging of the PCM and electricity generation. STEALS has inherent features that drive its cost-competitive scale to be much smaller than current commercial concentrating solar power (CSP) plants. Most obvious is modularity of the solid-state TEG, which favors smaller scales inmore » the kilowatt range as compared to CSP steam turbines, which are minimally 50 MWe for commercial power plants. Here, we present techno-economic and market analyses that show STEALS can be a cost-effective electricity-generating technology with particular appeal to small-scale microgrid applications. We evaluated levelized cost of energy (LCOE) for STEALS and for a comparable photovoltaic (PV) system with battery storage. For STEALS, we estimated capital costs and the LCOE as functions of the type of PCM including the use of recycled aluminum alloys, and evaluated the cost tradeoffs between plasma spray coatings and solution-based boron coatings that are applied to the wetted surfaces of the PCM subsystem. We developed a probabilistic cost model that accounts for uncertainties in the cost and performance inputs to the LCOE estimation. Our probabilistic model estimated LCOE for a 100-kWe STEALS system that had 5 hours of thermal storage and 8-10 hours of total daily power generation. For these cases, the solar multiple for the heliostat field varied between 1.12 and 1.5. We identified microgrids as a likely market for the STEALS system. We characterized microgrid markets in terms of nominal power, dispatchability, geographic location, and customer type, and specified additional features for STEALS that are needed to meet the needs of this growing power market.« less

Commercialization of NESSUS: Status

NASA Technical Reports Server (NTRS)

Thacker, Ben H.; Millwater, Harry R.

1991-01-01

A plan was initiated in 1988 to commercialize the Numerical Evaluation of Stochastic Structures Under Stress (NESSUS) probabilistic structural analysis software. The goal of the on-going commercialization effort is to begin the transfer of Probabilistic Structural Analysis Method (PSAM) developed technology into industry and to develop additional funding resources in the general area of structural reliability. The commercialization effort is summarized. The SwRI NESSUS Software System is a general purpose probabilistic finite element computer program using state of the art methods for predicting stochastic structural response due to random loads, material properties, part geometry, and boundary conditions. NESSUS can be used to assess structural reliability, to compute probability of failure, to rank the input random variables by importance, and to provide a more cost effective design than traditional methods. The goal is to develop a general probabilistic structural analysis methodology to assist in the certification of critical components in the next generation Space Shuttle Main Engine.

Effects of sample survey design on the accuracy of classification tree models in species distribution models

Treesearch

Thomas C. Edwards; D. Richard Cutler; Niklaus E. Zimmermann; Linda Geiser; Gretchen G. Moisen

2006-01-01

We evaluated the effects of probabilistic (hereafter DESIGN) and non-probabilistic (PURPOSIVE) sample surveys on resultant classification tree models for predicting the presence of four lichen species in the Pacific Northwest, USA. Models derived from both survey forms were assessed using an independent data set (EVALUATION). Measures of accuracy as gauged by...

Quantum probabilistic logic programming

NASA Astrophysics Data System (ADS)

Balu, Radhakrishnan

2015-05-01

We describe a quantum mechanics based logic programming language that supports Horn clauses, random variables, and covariance matrices to express and solve problems in probabilistic logic. The Horn clauses of the language wrap random variables, including infinite valued, to express probability distributions and statistical correlations, a powerful feature to capture relationship between distributions that are not independent. The expressive power of the language is based on a mechanism to implement statistical ensembles and to solve the underlying SAT instances using quantum mechanical machinery. We exploit the fact that classical random variables have quantum decompositions to build the Horn clauses. We establish the semantics of the language in a rigorous fashion by considering an existing probabilistic logic language called PRISM with classical probability measures defined on the Herbrand base and extending it to the quantum context. In the classical case H-interpretations form the sample space and probability measures defined on them lead to consistent definition of probabilities for well formed formulae. In the quantum counterpart, we define probability amplitudes on Hinterpretations facilitating the model generations and verifications via quantum mechanical superpositions and entanglements. We cast the well formed formulae of the language as quantum mechanical observables thus providing an elegant interpretation for their probabilities. We discuss several examples to combine statistical ensembles and predicates of first order logic to reason with situations involving uncertainty.

Modelling default and likelihood reasoning as probabilistic reasoning

NASA Technical Reports Server (NTRS)

Buntine, Wray

1990-01-01

A probabilistic analysis of plausible reasoning about defaults and about likelihood is presented. Likely and by default are in fact treated as duals in the same sense as possibility and necessity. To model these four forms probabilistically, a qualitative default probabilistic (QDP) logic and its quantitative counterpart DP are derived that allow qualitative and corresponding quantitative reasoning. Consistency and consequent results for subsets of the logics are given that require at most a quadratic number of satisfiability tests in the underlying propositional logic. The quantitative logic shows how to track the propagation error inherent in these reasoning forms. The methodology and sound framework of the system highlights their approximate nature, the dualities, and the need for complementary reasoning about relevance.

Probabilistic estimation of residential air exchange rates for population-based human exposure modeling

EPA Science Inventory

Residential air exchange rates (AERs) are a key determinant in the infiltration of ambient air pollution indoors. Population-based human exposure models using probabilistic approaches to estimate personal exposure to air pollutants have relied on input distributions from AER meas...

From information processing to decisions: Formalizing and comparing psychologically plausible choice models.

PubMed

Heck, Daniel W; Hilbig, Benjamin E; Moshagen, Morten

2017-08-01

Decision strategies explain how people integrate multiple sources of information to make probabilistic inferences. In the past decade, increasingly sophisticated methods have been developed to determine which strategy explains decision behavior best. We extend these efforts to test psychologically more plausible models (i.e., strategies), including a new, probabilistic version of the take-the-best (TTB) heuristic that implements a rank order of error probabilities based on sequential processing. Within a coherent statistical framework, deterministic and probabilistic versions of TTB and other strategies can directly be compared using model selection by minimum description length or the Bayes factor. In an experiment with inferences from given information, only three of 104 participants were best described by the psychologically plausible, probabilistic version of TTB. Similar as in previous studies, most participants were classified as users of weighted-additive, a strategy that integrates all available information and approximates rational decisions. Copyright © 2017 Elsevier Inc. All rights reserved.

Modeling Uncertainties in EEG Microstates: Analysis of Real and Imagined Motor Movements Using Probabilistic Clustering-Driven Training of Probabilistic Neural Networks.

PubMed

Dinov, Martin; Leech, Robert

2017-01-01

Part of the process of EEG microstate estimation involves clustering EEG channel data at the global field power (GFP) maxima, very commonly using a modified K-means approach. Clustering has also been done deterministically, despite there being uncertainties in multiple stages of the microstate analysis, including the GFP peak definition, the clustering itself and in the post-clustering assignment of microstates back onto the EEG timecourse of interest. We perform a fully probabilistic microstate clustering and labeling, to account for these sources of uncertainty using the closest probabilistic analog to KM called Fuzzy C-means (FCM). We train softmax multi-layer perceptrons (MLPs) using the KM and FCM-inferred cluster assignments as target labels, to then allow for probabilistic labeling of the full EEG data instead of the usual correlation-based deterministic microstate label assignment typically used. We assess the merits of the probabilistic analysis vs. the deterministic approaches in EEG data recorded while participants perform real or imagined motor movements from a publicly available data set of 109 subjects. Though FCM group template maps that are almost topographically identical to KM were found, there is considerable uncertainty in the subsequent assignment of microstate labels. In general, imagined motor movements are less predictable on a time point-by-time point basis, possibly reflecting the more exploratory nature of the brain state during imagined, compared to during real motor movements. We find that some relationships may be more evident using FCM than using KM and propose that future microstate analysis should preferably be performed probabilistically rather than deterministically, especially in situations such as with brain computer interfaces, where both training and applying models of microstates need to account for uncertainty. Probabilistic neural network-driven microstate assignment has a number of advantages that we have discussed, which are likely to be further developed and exploited in future studies. In conclusion, probabilistic clustering and a probabilistic neural network-driven approach to microstate analysis is likely to better model and reveal details and the variability hidden in current deterministic and binarized microstate assignment and analyses.

Modeling Uncertainties in EEG Microstates: Analysis of Real and Imagined Motor Movements Using Probabilistic Clustering-Driven Training of Probabilistic Neural Networks

PubMed Central

Dinov, Martin; Leech, Robert

2017-01-01

Part of the process of EEG microstate estimation involves clustering EEG channel data at the global field power (GFP) maxima, very commonly using a modified K-means approach. Clustering has also been done deterministically, despite there being uncertainties in multiple stages of the microstate analysis, including the GFP peak definition, the clustering itself and in the post-clustering assignment of microstates back onto the EEG timecourse of interest. We perform a fully probabilistic microstate clustering and labeling, to account for these sources of uncertainty using the closest probabilistic analog to KM called Fuzzy C-means (FCM). We train softmax multi-layer perceptrons (MLPs) using the KM and FCM-inferred cluster assignments as target labels, to then allow for probabilistic labeling of the full EEG data instead of the usual correlation-based deterministic microstate label assignment typically used. We assess the merits of the probabilistic analysis vs. the deterministic approaches in EEG data recorded while participants perform real or imagined motor movements from a publicly available data set of 109 subjects. Though FCM group template maps that are almost topographically identical to KM were found, there is considerable uncertainty in the subsequent assignment of microstate labels. In general, imagined motor movements are less predictable on a time point-by-time point basis, possibly reflecting the more exploratory nature of the brain state during imagined, compared to during real motor movements. We find that some relationships may be more evident using FCM than using KM and propose that future microstate analysis should preferably be performed probabilistically rather than deterministically, especially in situations such as with brain computer interfaces, where both training and applying models of microstates need to account for uncertainty. Probabilistic neural network-driven microstate assignment has a number of advantages that we have discussed, which are likely to be further developed and exploited in future studies. In conclusion, probabilistic clustering and a probabilistic neural network-driven approach to microstate analysis is likely to better model and reveal details and the variability hidden in current deterministic and binarized microstate assignment and analyses. PMID:29163110

ProbFold: a probabilistic method for integration of probing data in RNA secondary structure prediction.

PubMed

Sahoo, Sudhakar; Świtnicki, Michał P; Pedersen, Jakob Skou

2016-09-01

Recently, new RNA secondary structure probing techniques have been developed, including Next Generation Sequencing based methods capable of probing transcriptome-wide. These techniques hold great promise for improving structure prediction accuracy. However, each new data type comes with its own signal properties and biases, which may even be experiment specific. There is therefore a growing need for RNA structure prediction methods that can be automatically trained on new data types and readily extended to integrate and fully exploit multiple types of data. Here, we develop and explore a modular probabilistic approach for integrating probing data in RNA structure prediction. It can be automatically trained given a set of known structures with probing data. The approach is demonstrated on SHAPE datasets, where we evaluate and selectively model specific correlations. The approach often makes superior use of the probing data signal compared to other methods. We illustrate the use of ProbFold on multiple data types using both simulations and a small set of structures with both SHAPE, DMS and CMCT data. Technically, the approach combines stochastic context-free grammars (SCFGs) with probabilistic graphical models. This approach allows rapid adaptation and integration of new probing data types. ProbFold is implemented in C ++. Models are specified using simple textual formats. Data reformatting is done using separate C ++ programs. Source code, statically compiled binaries for x86 Linux machines, C ++ programs, example datasets and a tutorial is available from http://moma.ki.au.dk/prj/probfold/ : jakob.skou@clin.au.dk Supplementary data are available at Bioinformatics online. © The Author 2016. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.

Assessment of SWE data assimilation for ensemble streamflow predictions

NASA Astrophysics Data System (ADS)

Franz, Kristie J.; Hogue, Terri S.; Barik, Muhammad; He, Minxue

2014-11-01

An assessment of data assimilation (DA) for Ensemble Streamflow Prediction (ESP) using seasonal water supply hindcasting in the North Fork of the American River Basin (NFARB) and the National Weather Service (NWS) hydrologic forecast models is undertaken. Two parameter sets, one from the California Nevada River Forecast Center (RFC) and one from the Differential Evolution Adaptive Metropolis (DREAM) algorithm, are tested. For each parameter set, hindcasts are generated using initial conditions derived with and without the inclusion of a DA scheme that integrates snow water equivalent (SWE) observations. The DREAM-DA scenario uses an Integrated Uncertainty and Ensemble-based data Assimilation (ICEA) framework that also considers model and parameter uncertainty. Hindcasts are evaluated using deterministic and probabilistic forecast verification metrics. In general, the impact of DA on the skill of the seasonal water supply predictions is mixed. For deterministic (ensemble mean) predictions, the Percent Bias (PBias) is improved with integration of the DA. DREAM-DA and the RFC-DA have the lowest biases and the RFC-DA has the lowest Root Mean Squared Error (RMSE). However, the RFC and DREAM-DA have similar RMSE scores. For the probabilistic predictions, the RFC and DREAM have the highest Continuous Ranked Probability Skill Scores (CRPSS) and the RFC has the best discrimination for low flows. Reliability results are similar between the non-DA and DA tests and the DREAM and DREAM-DA have better reliability than the RFC and RFC-DA for forecast dates February 1 and later. Despite producing improved streamflow simulations in previous studies, the hindcast analysis suggests that the DA method tested may not result in obvious improvements in streamflow forecasts. We advocate that integration of hindcasting and probabilistic metrics provides more rigorous insight on model performance for forecasting applications, such as in this study.

Multi-Topic Tracking Model for dynamic social network

NASA Astrophysics Data System (ADS)

Li, Yuhua; Liu, Changzheng; Zhao, Ming; Li, Ruixuan; Xiao, Hailing; Wang, Kai; Zhang, Jun

2016-07-01

The topic tracking problem has attracted much attention in the last decades. However, existing approaches rarely consider network structures and textual topics together. In this paper, we propose a novel statistical model based on dynamic bayesian network, namely Multi-Topic Tracking Model for Dynamic Social Network (MTTD). It takes influence phenomenon, selection phenomenon, document generative process and the evolution of textual topics into account. Specifically, in our MTTD model, Gibbs Random Field is defined to model the influence of historical status of users in the network and the interdependency between them in order to consider the influence phenomenon. To address the selection phenomenon, a stochastic block model is used to model the link generation process based on the users' interests to topics. Probabilistic Latent Semantic Analysis (PLSA) is used to describe the document generative process according to the users' interests. Finally, the dependence on the historical topic status is also considered to ensure the continuity of the topic itself in topic evolution model. Expectation Maximization (EM) algorithm is utilized to estimate parameters in the proposed MTTD model. Empirical experiments on real datasets show that the MTTD model performs better than Popular Event Tracking (PET) and Dynamic Topic Model (DTM) in generalization performance, topic interpretability performance, topic content evolution and topic popularity evolution performance.

Decomposing biodiversity data using the Latent Dirichlet Allocation model, a probabilistic multivariate statistical method

Treesearch

Denis Valle; Benjamin Baiser; Christopher W. Woodall; Robin Chazdon; Jerome Chave

2014-01-01

We propose a novel multivariate method to analyse biodiversity data based on the Latent Dirichlet Allocation (LDA) model. LDA, a probabilistic model, reduces assemblages to sets of distinct component communities. It produces easily interpretable results, can represent abrupt and gradual changes in composition, accommodates missing data and allows for coherent estimates...

Probabilistic Inference: Task Dependency and Individual Differences of Probability Weighting Revealed by Hierarchical Bayesian Modeling

PubMed Central

Boos, Moritz; Seer, Caroline; Lange, Florian; Kopp, Bruno

2016-01-01

Cognitive determinants of probabilistic inference were examined using hierarchical Bayesian modeling techniques. A classic urn-ball paradigm served as experimental strategy, involving a factorial two (prior probabilities) by two (likelihoods) design. Five computational models of cognitive processes were compared with the observed behavior. Parameter-free Bayesian posterior probabilities and parameter-free base rate neglect provided inadequate models of probabilistic inference. The introduction of distorted subjective probabilities yielded more robust and generalizable results. A general class of (inverted) S-shaped probability weighting functions had been proposed; however, the possibility of large differences in probability distortions not only across experimental conditions, but also across individuals, seems critical for the model's success. It also seems advantageous to consider individual differences in parameters of probability weighting as being sampled from weakly informative prior distributions of individual parameter values. Thus, the results from hierarchical Bayesian modeling converge with previous results in revealing that probability weighting parameters show considerable task dependency and individual differences. Methodologically, this work exemplifies the usefulness of hierarchical Bayesian modeling techniques for cognitive psychology. Theoretically, human probabilistic inference might be best described as the application of individualized strategic policies for Bayesian belief revision. PMID:27303323

Model fitting data from syllogistic reasoning experiments.

PubMed

Hattori, Masasi

2016-12-01

The data presented in this article are related to the research article entitled "Probabilistic representation in syllogistic reasoning: A theory to integrate mental models and heuristics" (M. Hattori, 2016) [1]. This article presents predicted data by three signature probabilistic models of syllogistic reasoning and model fitting results for each of a total of 12 experiments ( N =404) in the literature. Models are implemented in R, and their source code is also provided.

The Probability Heuristics Model of Syllogistic Reasoning.

ERIC Educational Resources Information Center

Chater, Nick; Oaksford, Mike

1999-01-01

Proposes a probability heuristic model for syllogistic reasoning and confirms the rationality of this heuristic by an analysis of the probabilistic validity of syllogistic reasoning that treats logical inference as a limiting case of probabilistic inference. Meta-analysis and two experiments involving 40 adult participants and using generalized…

Modeling Array Stations in SIG-VISA

NASA Astrophysics Data System (ADS)

Ding, N.; Moore, D.; Russell, S.

2013-12-01

We add support for array stations to SIG-VISA, a system for nuclear monitoring using probabilistic inference on seismic signals. Array stations comprise a large portion of the IMS network; they can provide increased sensitivity and more accurate directional information compared to single-component stations. Our existing model assumed that signals were independent at each station, which is false when lots of stations are close together, as in an array. The new model removes that assumption by jointly modeling signals across array elements. This is done by extending our existing Gaussian process (GP) regression models, also known as kriging, from a 3-dimensional single-component space of events to a 6-dimensional space of station-event pairs. For each array and each event attribute (including coda decay, coda height, amplitude transfer and travel time), we model the joint distribution across array elements using a Gaussian process that learns the correlation lengthscale across the array, thereby incorporating information of array stations into the probabilistic inference framework. To evaluate the effectiveness of our model, we perform ';probabilistic beamforming' on new events using our GP model, i.e., we compute the event azimuth having highest posterior probability under the model, conditioned on the signals at array elements. We compare the results from our probabilistic inference model to the beamforming currently performed by IMS station processing.

Development of probabilistic regional climate scenario in East Asia

NASA Astrophysics Data System (ADS)

Dairaku, K.; Ueno, G.; Ishizaki, N. N.

2015-12-01

Climate information and services for Impacts, Adaptation and Vulnerability (IAV) Assessments are of great concern. In order to develop probabilistic regional climate information that represents the uncertainty in climate scenario experiments in East Asia (CORDEX-EA and Japan), the probability distribution of 2m air temperature was estimated by using developed regression model. The method can be easily applicable to other regions and other physical quantities, and also to downscale to finer-scale dependent on availability of observation dataset. Probabilistic climate information in present (1969-1998) and future (2069-2098) climate was developed using CMIP3 SRES A1b scenarios 21 models and the observation data (CRU_TS3.22 & University of Delaware in CORDEX-EA, NIAES AMeDAS mesh data in Japan). The prototype of probabilistic information in CORDEX-EA and Japan represent the quantified structural uncertainties of multi-model ensemble experiments of climate change scenarios. Appropriate combination of statistical methods and optimization of climate ensemble experiments using multi-General Circulation Models (GCMs) and multi-regional climate models (RCMs) ensemble downscaling experiments are investigated.

Probabilistic inversion of expert assessments to inform projections about Antarctic ice sheet responses.

PubMed

Fuller, Robert William; Wong, Tony E; Keller, Klaus

2017-01-01

The response of the Antarctic ice sheet (AIS) to changing global temperatures is a key component of sea-level projections. Current projections of the AIS contribution to sea-level changes are deeply uncertain. This deep uncertainty stems, in part, from (i) the inability of current models to fully resolve key processes and scales, (ii) the relatively sparse available data, and (iii) divergent expert assessments. One promising approach to characterizing the deep uncertainty stemming from divergent expert assessments is to combine expert assessments, observations, and simple models by coupling probabilistic inversion and Bayesian inversion. Here, we present a proof-of-concept study that uses probabilistic inversion to fuse a simple AIS model and diverse expert assessments. We demonstrate the ability of probabilistic inversion to infer joint prior probability distributions of model parameters that are consistent with expert assessments. We then confront these inferred expert priors with instrumental and paleoclimatic observational data in a Bayesian inversion. These additional constraints yield tighter hindcasts and projections. We use this approach to quantify how the deep uncertainty surrounding expert assessments affects the joint probability distributions of model parameters and future projections.

A Semi-Automated Machine Learning Algorithm for Tree Cover Delineation from 1-m Naip Imagery Using a High Performance Computing Architecture

NASA Astrophysics Data System (ADS)

Basu, S.; Ganguly, S.; Nemani, R. R.; Mukhopadhyay, S.; Milesi, C.; Votava, P.; Michaelis, A.; Zhang, G.; Cook, B. D.; Saatchi, S. S.; Boyda, E.

2014-12-01

Accurate tree cover delineation is a useful instrument in the derivation of Above Ground Biomass (AGB) density estimates from Very High Resolution (VHR) satellite imagery data. Numerous algorithms have been designed to perform tree cover delineation in high to coarse resolution satellite imagery, but most of them do not scale to terabytes of data, typical in these VHR datasets. In this paper, we present an automated probabilistic framework for the segmentation and classification of 1-m VHR data as obtained from the National Agriculture Imagery Program (NAIP) for deriving tree cover estimates for the whole of Continental United States, using a High Performance Computing Architecture. The results from the classification and segmentation algorithms are then consolidated into a structured prediction framework using a discriminative undirected probabilistic graphical model based on Conditional Random Field (CRF), which helps in capturing the higher order contextual dependencies between neighboring pixels. Once the final probability maps are generated, the framework is updated and re-trained by incorporating expert knowledge through the relabeling of misclassified image patches. This leads to a significant improvement in the true positive rates and reduction in false positive rates. The tree cover maps were generated for the state of California, which covers a total of 11,095 NAIP tiles and spans a total geographical area of 163,696 sq. miles. Our framework produced correct detection rates of around 85% for fragmented forests and 70% for urban tree cover areas, with false positive rates lower than 3% for both regions. Comparative studies with the National Land Cover Data (NLCD) algorithm and the LiDAR high-resolution canopy height model shows the effectiveness of our algorithm in generating accurate high-resolution tree cover maps.

Predictive coarse-graining

NASA Astrophysics Data System (ADS)

Schöberl, Markus; Zabaras, Nicholas; Koutsourelakis, Phaedon-Stelios

2017-03-01

We propose a data-driven, coarse-graining formulation in the context of equilibrium statistical mechanics. In contrast to existing techniques which are based on a fine-to-coarse map, we adopt the opposite strategy by prescribing a probabilistic coarse-to-fine map. This corresponds to a directed probabilistic model where the coarse variables play the role of latent generators of the fine scale (all-atom) data. From an information-theoretic perspective, the framework proposed provides an improvement upon the relative entropy method [1] and is capable of quantifying the uncertainty due to the information loss that unavoidably takes place during the coarse-graining process. Furthermore, it can be readily extended to a fully Bayesian model where various sources of uncertainties are reflected in the posterior of the model parameters. The latter can be used to produce not only point estimates of fine-scale reconstructions or macroscopic observables, but more importantly, predictive posterior distributions on these quantities. Predictive posterior distributions reflect the confidence of the model as a function of the amount of data and the level of coarse-graining. The issues of model complexity and model selection are seamlessly addressed by employing a hierarchical prior that favors the discovery of sparse solutions, revealing the most prominent features in the coarse-grained model. A flexible and parallelizable Monte Carlo - Expectation-Maximization (MC-EM) scheme is proposed for carrying out inference and learning tasks. A comparative assessment of the proposed methodology is presented for a lattice spin system and the SPC/E water model.

Predictive coarse-graining

DOE Office of Scientific and Technical Information (OSTI.GOV)

Schöberl, Markus, E-mail: m.schoeberl@tum.de; Zabaras, Nicholas; Department of Aerospace and Mechanical Engineering, University of Notre Dame, 365 Fitzpatrick Hall, Notre Dame, IN 46556

We propose a data-driven, coarse-graining formulation in the context of equilibrium statistical mechanics. In contrast to existing techniques which are based on a fine-to-coarse map, we adopt the opposite strategy by prescribing a probabilistic coarse-to-fine map. This corresponds to a directed probabilistic model where the coarse variables play the role of latent generators of the fine scale (all-atom) data. From an information-theoretic perspective, the framework proposed provides an improvement upon the relative entropy method and is capable of quantifying the uncertainty due to the information loss that unavoidably takes place during the coarse-graining process. Furthermore, it can be readily extendedmore » to a fully Bayesian model where various sources of uncertainties are reflected in the posterior of the model parameters. The latter can be used to produce not only point estimates of fine-scale reconstructions or macroscopic observables, but more importantly, predictive posterior distributions on these quantities. Predictive posterior distributions reflect the confidence of the model as a function of the amount of data and the level of coarse-graining. The issues of model complexity and model selection are seamlessly addressed by employing a hierarchical prior that favors the discovery of sparse solutions, revealing the most prominent features in the coarse-grained model. A flexible and parallelizable Monte Carlo – Expectation–Maximization (MC-EM) scheme is proposed for carrying out inference and learning tasks. A comparative assessment of the proposed methodology is presented for a lattice spin system and the SPC/E water model.« less

Probabilistic seismic history matching using binary images

NASA Astrophysics Data System (ADS)

Davolio, Alessandra; Schiozer, Denis Jose

2018-02-01

Currently, the goal of history-matching procedures is not only to provide a model matching any observed data but also to generate multiple matched models to properly handle uncertainties. One such approach is a probabilistic history-matching methodology based on the discrete Latin Hypercube sampling algorithm, proposed in previous works, which was particularly efficient for matching well data (production rates and pressure). 4D seismic (4DS) data have been increasingly included into history-matching procedures. A key issue in seismic history matching (SHM) is to transfer data into a common domain: impedance, amplitude or pressure, and saturation. In any case, seismic inversions and/or modeling are required, which can be time consuming. An alternative to avoid these procedures is using binary images in SHM as they allow the shape, rather than the physical values, of observed anomalies to be matched. This work presents the incorporation of binary images in SHM within the aforementioned probabilistic history matching. The application was performed with real data from a segment of the Norne benchmark case that presents strong 4D anomalies, including softening signals due to pressure build up. The binary images are used to match the pressurized zones observed in time-lapse data. Three history matchings were conducted using: only well data, well and 4DS data, and only 4DS. The methodology is very flexible and successfully utilized the addition of binary images for seismic objective functions. Results proved the good convergence of the method in few iterations for all three cases. The matched models of the first two cases provided the best results, with similar well matching quality. The second case provided models presenting pore pressure changes according to the expected dynamic behavior (pressurized zones) observed on 4DS data. The use of binary images in SHM is relatively new with few examples in the literature. This work enriches this discussion by presenting a new application to match pressure in a reservoir segment with complex pressure behavior.

Improved reliability of wind turbine towers with active tuned mass dampers (ATMDs)

NASA Astrophysics Data System (ADS)

Fitzgerald, Breiffni; Sarkar, Saptarshi; Staino, Andrea

2018-04-01

Modern multi-megawatt wind turbines are composed of slender, flexible, and lightly damped blades and towers. These components exhibit high susceptibility to wind-induced vibrations. As the size, flexibility and cost of the towers have increased in recent years, the need to protect these structures against damage induced by turbulent aerodynamic loading has become apparent. This paper combines structural dynamic models and probabilistic assessment tools to demonstrate improvements in structural reliability when modern wind turbine towers are equipped with active tuned mass dampers (ATMDs). This study proposes a multi-modal wind turbine model for wind turbine control design and analysis. This study incorporates an ATMD into the tower of this model. The model is subjected to stochastically generated wind loads of varying speeds to develop wind-induced probabilistic demand models for towers of modern multi-megawatt wind turbines under structural uncertainty. Numerical simulations have been carried out to ascertain the effectiveness of the active control system to improve the structural performance of the wind turbine and its reliability. The study constructs fragility curves, which illustrate reductions in the vulnerability of towers to wind loading owing to the inclusion of the damper. Results show that the active controller is successful in increasing the reliability of the tower responses. According to the analysis carried out in this paper, a strong reduction of the probability of exceeding a given displacement at the rated wind speed has been observed.

a Probabilistic Embedding Clustering Method for Urban Structure Detection

NASA Astrophysics Data System (ADS)

Lin, X.; Li, H.; Zhang, Y.; Gao, L.; Zhao, L.; Deng, M.

2017-09-01

Urban structure detection is a basic task in urban geography. Clustering is a core technology to detect the patterns of urban spatial structure, urban functional region, and so on. In big data era, diverse urban sensing datasets recording information like human behaviour and human social activity, suffer from complexity in high dimension and high noise. And unfortunately, the state-of-the-art clustering methods does not handle the problem with high dimension and high noise issues concurrently. In this paper, a probabilistic embedding clustering method is proposed. Firstly, we come up with a Probabilistic Embedding Model (PEM) to find latent features from high dimensional urban sensing data by "learning" via probabilistic model. By latent features, we could catch essential features hidden in high dimensional data known as patterns; with the probabilistic model, we can also reduce uncertainty caused by high noise. Secondly, through tuning the parameters, our model could discover two kinds of urban structure, the homophily and structural equivalence, which means communities with intensive interaction or in the same roles in urban structure. We evaluated the performance of our model by conducting experiments on real-world data and experiments with real data in Shanghai (China) proved that our method could discover two kinds of urban structure, the homophily and structural equivalence, which means clustering community with intensive interaction or under the same roles in urban space.

Visualising probabilistic flood forecast information: expert preferences and perceptions of best practice in uncertainty communication

NASA Astrophysics Data System (ADS)

Pappenberger, F.; Stephens, E. M.; Thielen, J.; Salomon, P.; Demeritt, D.; van Andel, S.; Wetterhall, F.; Alfieri, L.

2011-12-01

The aim of this paper is to understand and to contribute to improved communication of the probabilistic flood forecasts generated by Hydrological Ensemble Prediction Systems (HEPS) with particular focus on the inter expert communication. Different users are likely to require different kinds of information from HEPS and thus different visualizations. The perceptions of this expert group are important both because they are the designers and primary users of existing HEPS. Nevertheless, they have sometimes resisted the release of uncertainty information to the general public because of doubts about whether it can be successfully communicated in ways that would be readily understood to non-experts. In this paper we explore the strengths and weaknesses of existing HEPS visualization methods and thereby formulate some wider recommendations about best practice for HEPS visualization and communication. We suggest that specific training on probabilistic forecasting would foster use of probabilistic forecasts with a wider range of applications. The result of a case study exercise showed that there is no overarching agreement between experts on how to display probabilistic forecasts and what they consider essential information that should accompany plots and diagrams. In this paper we propose a list of minimum properties that, if consistently displayed with probabilistic forecasts, would make the products more easily understandable.

Composite load spectra for select space propulsion structural components

NASA Technical Reports Server (NTRS)

Newell, J. F.; Ho, H. W.; Kurth, R. E.

1991-01-01

The work performed to develop composite load spectra (CLS) for the Space Shuttle Main Engine (SSME) using probabilistic methods. The three methods were implemented to be the engine system influence model. RASCAL was chosen to be the principal method as most component load models were implemented with the method. Validation of RASCAL was performed. High accuracy comparable to the Monte Carlo method can be obtained if a large enough bin size is used. Generic probabilistic models were developed and implemented for load calculations using the probabilistic methods discussed above. Each engine mission, either a real fighter or a test, has three mission phases: the engine start transient phase, the steady state phase, and the engine cut off transient phase. Power level and engine operating inlet conditions change during a mission. The load calculation module provides the steady-state and quasi-steady state calculation procedures with duty-cycle-data option. The quasi-steady state procedure is for engine transient phase calculations. In addition, a few generic probabilistic load models were also developed for specific conditions. These include the fixed transient spike model, the poison arrival transient spike model, and the rare event model. These generic probabilistic load models provide sufficient latitude for simulating loads with specific conditions. For SSME components, turbine blades, transfer ducts, LOX post, and the high pressure oxidizer turbopump (HPOTP) discharge duct were selected for application of the CLS program. They include static pressure loads and dynamic pressure loads for all four components, centrifugal force for the turbine blade, temperatures of thermal loads for all four components, and structural vibration loads for the ducts and LOX posts.

Cost-effectiveness of drug-eluting stents versus bare-metal stents in patients undergoing percutaneous coronary intervention.

PubMed

Baschet, Louise; Bourguignon, Sandrine; Marque, Sébastien; Durand-Zaleski, Isabelle; Teiger, Emmanuel; Wilquin, Fanny; Levesque, Karine

2016-01-01

To determine the cost-effectiveness of drug-eluting stents (DES) compared with bare-metal stents (BMS) in patients requiring a percutaneous coronary intervention in France, using a recent meta-analysis including second-generation DES. A cost-effectiveness analysis was performed in the French National Health Insurance setting. Effectiveness settings were taken from a meta-analysis of 117 762 patient-years with 76 randomised trials. The main effectiveness criterion was major cardiac event-free survival. Effectiveness and costs were modelled over a 5-year horizon using a three-state Markov model. Incremental cost-effectiveness ratios and a cost-effectiveness acceptability curve were calculated for a range of thresholds for willingness to pay per year without major cardiac event gain. Deterministic and probabilistic sensitivity analyses were performed. Base case results demonstrated that DES are dominant over BMS, with an increase in event-free survival and a cost-reduction of €184, primarily due to a diminution of second revascularisations, and an absence of myocardial infarction and stent thrombosis. These results are robust for uncertainty on one-way deterministic and probabilistic sensitivity analyses. Using a cost-effectiveness threshold of €7000 per major cardiac event-free year gained, DES has a >95% probability of being cost-effective versus BMS. Following DES price decrease, new-generation DES development and taking into account recent meta-analyses results, the DES can now be considered cost-effective regardless of selective indication in France, according to European recommendations.

Three-body system metaphor for the two-slit experiment and Escherichia coli lactose-glucose metabolism.

PubMed

Asano, Masanari; Khrennikov, Andrei; Ohya, Masanori; Tanaka, Yoshiharu; Yamato, Ichiro

2016-05-28

We compare the contextual probabilistic structures of the seminal two-slit experiment (quantum interference experiment), the system of three interacting bodies andEscherichia colilactose-glucose metabolism. We show that they have the same non-Kolmogorov probabilistic structure resulting from multi-contextuality. There are plenty of statistical data with non-Kolmogorov features; in particular, the probabilistic behaviour of neither quantum nor biological systems can be described classically. Biological systems (even cells and proteins) are macroscopic systems and one may try to present a more detailed model of interactions in such systems that lead to quantum-like probabilistic behaviour. The system of interactions between three bodies is one of the simplest metaphoric examples for such interactions. By proceeding further in this way (by playing withn-body systems) we shall be able to find metaphoric mechanical models for complex bio-interactions, e.g. signalling between cells, leading to non-Kolmogorov probabilistic data. © 2016 The Author(s).

Scalable DB+IR Technology: Processing Probabilistic Datalog with HySpirit.

PubMed

Frommholz, Ingo; Roelleke, Thomas

2016-01-01

Probabilistic Datalog (PDatalog, proposed in 1995) is a probabilistic variant of Datalog and a nice conceptual idea to model Information Retrieval in a logical, rule-based programming paradigm. Making PDatalog work in real-world applications requires more than probabilistic facts and rules, and the semantics associated with the evaluation of the programs. We report in this paper some of the key features of the HySpirit system required to scale the execution of PDatalog programs. Firstly, there is the requirement to express probability estimation in PDatalog. Secondly, fuzzy-like predicates are required to model vague predicates (e.g. vague match of attributes such as age or price). Thirdly, to handle large data sets there are scalability issues to be addressed, and therefore, HySpirit provides probabilistic relational indexes and parallel and distributed processing . The main contribution of this paper is a consolidated view on the methods of the HySpirit system to make PDatalog applicable in real-scale applications that involve a wide range of requirements typical for data (information) management and analysis.

Three-body system metaphor for the two-slit experiment and Escherichia coli lactose–glucose metabolism

PubMed Central

Asano, Masanari; Ohya, Masanori; Yamato, Ichiro

2016-01-01

We compare the contextual probabilistic structures of the seminal two-slit experiment (quantum interference experiment), the system of three interacting bodies and Escherichia coli lactose–glucose metabolism. We show that they have the same non-Kolmogorov probabilistic structure resulting from multi-contextuality. There are plenty of statistical data with non-Kolmogorov features; in particular, the probabilistic behaviour of neither quantum nor biological systems can be described classically. Biological systems (even cells and proteins) are macroscopic systems and one may try to present a more detailed model of interactions in such systems that lead to quantum-like probabilistic behaviour. The system of interactions between three bodies is one of the simplest metaphoric examples for such interactions. By proceeding further in this way (by playing with n-body systems) we shall be able to find metaphoric mechanical models for complex bio-interactions, e.g. signalling between cells, leading to non-Kolmogorov probabilistic data. PMID:27091163

Modelling ecosystem service flows under uncertainty with stochiastic SPAN

USGS Publications Warehouse

Johnson, Gary W.; Snapp, Robert R.; Villa, Ferdinando; Bagstad, Kenneth J.

2012-01-01

Ecosystem service models are increasingly in demand for decision making. However, the data required to run these models are often patchy, missing, outdated, or untrustworthy. Further, communication of data and model uncertainty to decision makers is often either absent or unintuitive. In this work, we introduce a systematic approach to addressing both the data gap and the difficulty in communicating uncertainty through a stochastic adaptation of the Service Path Attribution Networks (SPAN) framework. The SPAN formalism assesses ecosystem services through a set of up to 16 maps, which characterize the services in a study area in terms of flow pathways between ecosystems and human beneficiaries. Although the SPAN algorithms were originally defined deterministically, we present them here in a stochastic framework which combines probabilistic input data with a stochastic transport model in order to generate probabilistic spatial outputs. This enables a novel feature among ecosystem service models: the ability to spatially visualize uncertainty in the model results. The stochastic SPAN model can analyze areas where data limitations are prohibitive for deterministic models. Greater uncertainty in the model inputs (including missing data) should lead to greater uncertainty expressed in the model’s output distributions. By using Bayesian belief networks to fill data gaps and expert-provided trust assignments to augment untrustworthy or outdated information, we can account for uncertainty in input data, producing a model that is still able to run and provide information where strictly deterministic models could not. Taken together, these attributes enable more robust and intuitive modelling of ecosystem services under uncertainty.

Trait-Dependent Biogeography: (Re)Integrating Biology into Probabilistic Historical Biogeographical Models.

PubMed

Sukumaran, Jeet; Knowles, L Lacey

2018-06-01

The development of process-based probabilistic models for historical biogeography has transformed the field by grounding it in modern statistical hypothesis testing. However, most of these models abstract away biological differences, reducing species to interchangeable lineages. We present here the case for reintegration of biology into probabilistic historical biogeographical models, allowing a broader range of questions about biogeographical processes beyond ancestral range estimation or simple correlation between a trait and a distribution pattern, as well as allowing us to assess how inferences about ancestral ranges themselves might be impacted by differential biological traits. We show how new approaches to inference might cope with the computational challenges resulting from the increased complexity of these trait-based historical biogeographical models. Copyright © 2018 Elsevier Ltd. All rights reserved.

Comparison of Four Probabilistic Models (CARES, Calendex, ConsEspo, SHEDS) to Estimate Aggregate Residential Exposures to Pesticides

EPA Science Inventory

Two deterministic models (US EPA’s Office of Pesticide Programs Residential Standard Operating Procedures (OPP Residential SOPs) and Draft Protocol for Measuring Children’s Non-Occupational Exposure to Pesticides by all Relevant Pathways (Draft Protocol)) and four probabilistic mo...

EXPERIENCES WITH USING PROBABILISTIC EXPOSURE ANALYSIS METHODS IN THE U.S. EPA

EPA Science Inventory

Over the past decade various Offices and Programs within the U.S. EPA have either initiated or increased the development and application of probabilistic exposure analysis models. These models have been applied to a broad range of research or regulatory problems in EPA, such as e...

Structural reliability assessment capability in NESSUS

NASA Technical Reports Server (NTRS)

Millwater, H.; Wu, Y.-T.

1992-01-01

The principal capabilities of NESSUS (Numerical Evaluation of Stochastic Structures Under Stress), an advanced computer code developed for probabilistic structural response analysis, are reviewed, and its structural reliability assessed. The code combines flexible structural modeling tools with advanced probabilistic algorithms in order to compute probabilistic structural response and resistance, component reliability and risk, and system reliability and risk. An illustrative numerical example is presented.

Structural reliability assessment capability in NESSUS

NASA Astrophysics Data System (ADS)

Millwater, H.; Wu, Y.-T.

1992-07-01

The principal capabilities of NESSUS (Numerical Evaluation of Stochastic Structures Under Stress), an advanced computer code developed for probabilistic structural response analysis, are reviewed, and its structural reliability assessed. The code combines flexible structural modeling tools with advanced probabilistic algorithms in order to compute probabilistic structural response and resistance, component reliability and risk, and system reliability and risk. An illustrative numerical example is presented.

Distributed collaborative probabilistic design of multi-failure structure with fluid-structure interaction using fuzzy neural network of regression

NASA Astrophysics Data System (ADS)

Song, Lu-Kai; Wen, Jie; Fei, Cheng-Wei; Bai, Guang-Chen

2018-05-01

To improve the computing efficiency and precision of probabilistic design for multi-failure structure, a distributed collaborative probabilistic design method-based fuzzy neural network of regression (FR) (called as DCFRM) is proposed with the integration of distributed collaborative response surface method and fuzzy neural network regression model. The mathematical model of DCFRM is established and the probabilistic design idea with DCFRM is introduced. The probabilistic analysis of turbine blisk involving multi-failure modes (deformation failure, stress failure and strain failure) was investigated by considering fluid-structure interaction with the proposed method. The distribution characteristics, reliability degree, and sensitivity degree of each failure mode and overall failure mode on turbine blisk are obtained, which provides a useful reference for improving the performance and reliability of aeroengine. Through the comparison of methods shows that the DCFRM reshapes the probability of probabilistic analysis for multi-failure structure and improves the computing efficiency while keeping acceptable computational precision. Moreover, the proposed method offers a useful insight for reliability-based design optimization of multi-failure structure and thereby also enriches the theory and method of mechanical reliability design.

A novel probabilistic framework for event-based speech recognition

NASA Astrophysics Data System (ADS)

Juneja, Amit; Espy-Wilson, Carol

2003-10-01

One of the reasons for unsatisfactory performance of the state-of-the-art automatic speech recognition (ASR) systems is the inferior acoustic modeling of low-level acoustic-phonetic information in the speech signal. An acoustic-phonetic approach to ASR, on the other hand, explicitly targets linguistic information in the speech signal, but such a system for continuous speech recognition (CSR) is not known to exist. A probabilistic and statistical framework for CSR based on the idea of the representation of speech sounds by bundles of binary valued articulatory phonetic features is proposed. Multiple probabilistic sequences of linguistically motivated landmarks are obtained using binary classifiers of manner phonetic features-syllabic, sonorant and continuant-and the knowledge-based acoustic parameters (APs) that are acoustic correlates of those features. The landmarks are then used for the extraction of knowledge-based APs for source and place phonetic features and their binary classification. Probabilistic landmark sequences are constrained using manner class language models for isolated or connected word recognition. The proposed method could overcome the disadvantages encountered by the early acoustic-phonetic knowledge-based systems that led the ASR community to switch to systems highly dependent on statistical pattern analysis methods and probabilistic language or grammar models.

Comparison and validation of shallow landslides susceptibility maps generated by bi-variate and multi-variate linear probabilistic GIS-based techniques. A case study from Ribeira Quente Valley (S. Miguel Island, Azores)

NASA Astrophysics Data System (ADS)

Marques, R.; Amaral, P.; Zêzere, J. L.; Queiroz, G.; Goulart, C.

2009-04-01

Slope instability research and susceptibility mapping is a fundamental component of hazard assessment and is of extreme importance for risk mitigation, land-use management and emergency planning. Landslide susceptibility zonation has been actively pursued during the last two decades and several methodologies are still being improved. Among all the methods presented in the literature, indirect quantitative probabilistic methods have been extensively used. In this work different linear probabilistic methods, both bi-variate and multi-variate (Informative Value, Fuzzy Logic, Weights of Evidence and Logistic Regression), were used for the computation of the spatial probability of landslide occurrence, using the pixel as mapping unit. The methods used are based on linear relationships between landslides and 9 considered conditioning factors (altimetry, slope angle, exposition, curvature, distance to streams, wetness index, contribution area, lithology and land-use). It was assumed that future landslides will be conditioned by the same factors as past landslides in the study area. The presented work was developed for Ribeira Quente Valley (S. Miguel Island, Azores), a study area of 9,5 km2, mainly composed of volcanic deposits (ash and pumice lapilli) produced by explosive eruptions in Furnas Volcano. This materials associated to the steepness of the slopes (38,9% of the area has slope angles higher than 35°, reaching a maximum of 87,5°), make the area very prone to landslide activity. A total of 1.495 shallow landslides were mapped (at 1:5.000 scale) and included in a GIS database. The total affected area is 401.744 m2 (4,5% of the study area). Most slope movements are translational slides frequently evolving into debris-flows. The landslides are elongated, with maximum length generally equivalent to the slope extent, and their width normally does not exceed 25 m. The failure depth rarely exceeds 1,5 m and the volume is usually smaller than 700 m3. For modelling purposes, the landslides were randomly divided in two sub-datasets: a modelling dataset with 748 events (2,2% of the study area) and a validation dataset with 747 events (2,3% of the study area). The susceptibility algorithms achieved with the different probabilistic techniques, were rated individually using success rate and prediction rate curves. The best model performance was obtained with the logistic regression, although the results from the different methods do not show significant differences neither in success nor in prediction rate curves. These evidences revealed that: (1) the modelling landslide dataset is representative of the entire landslide population characteristics; and (2) the increase of complexity and robustness in the probabilistic methodology did not produce a significant increase in success or prediction rates. Therefore, it was concluded that the resolution and quality of the input variables are much more important than the probabilistic model chosen to assess landslide susceptibility. This work was developed on the behalf of VOLCSOILRISK project (Volcanic Soils Geotechnical Characterization for Landslide Risk Mitigation), supported by Direcção Regional da Ciência e Tecnologia - Governo Regional dos Açores.

Probabilistic risk models for multiple disturbances: an example of forest insects and wildfires

Treesearch

Haiganoush K. Preisler; Alan A. Ager; Jane L. Hayes

2010-01-01

Building probabilistic risk models for highly random forest disturbances like wildfire and forest insect outbreaks is a challenging. Modeling the interactions among natural disturbances is even more difficult. In the case of wildfire and forest insects, we looked at the probability of a large fire given an insect outbreak and also the incidence of insect outbreaks...

A Markov model of the Indus script

PubMed Central

Rao, Rajesh P. N.; Yadav, Nisha; Vahia, Mayank N.; Joglekar, Hrishikesh; Adhikari, R.; Mahadevan, Iravatham

2009-01-01

Although no historical information exists about the Indus civilization (flourished ca. 2600–1900 B.C.), archaeologists have uncovered about 3,800 short samples of a script that was used throughout the civilization. The script remains undeciphered, despite a large number of attempts and claimed decipherments over the past 80 years. Here, we propose the use of probabilistic models to analyze the structure of the Indus script. The goal is to reveal, through probabilistic analysis, syntactic patterns that could point the way to eventual decipherment. We illustrate the approach using a simple Markov chain model to capture sequential dependencies between signs in the Indus script. The trained model allows new sample texts to be generated, revealing recurring patterns of signs that could potentially form functional subunits of a possible underlying language. The model also provides a quantitative way of testing whether a particular string belongs to the putative language as captured by the Markov model. Application of this test to Indus seals found in Mesopotamia and other sites in West Asia reveals that the script may have been used to express different content in these regions. Finally, we show how missing, ambiguous, or unreadable signs on damaged objects can be filled in with most likely predictions from the model. Taken together, our results indicate that the Indus script exhibits rich synactic structure and the ability to represent diverse content. both of which are suggestive of a linguistic writing system rather than a nonlinguistic symbol system. PMID:19666571

A Probabilistic Asteroid Impact Risk Model

NASA Technical Reports Server (NTRS)

Mathias, Donovan L.; Wheeler, Lorien F.; Dotson, Jessie L.

2016-01-01

Asteroid threat assessment requires the quantification of both the impact likelihood and resulting consequence across the range of possible events. This paper presents a probabilistic asteroid impact risk (PAIR) assessment model developed for this purpose. The model incorporates published impact frequency rates with state-of-the-art consequence assessment tools, applied within a Monte Carlo framework that generates sets of impact scenarios from uncertain parameter distributions. Explicit treatment of atmospheric entry is included to produce energy deposition rates that account for the effects of thermal ablation and object fragmentation. These energy deposition rates are used to model the resulting ground damage, and affected populations are computed for the sampled impact locations. The results for each scenario are aggregated into a distribution of potential outcomes that reflect the range of uncertain impact parameters, population densities, and strike probabilities. As an illustration of the utility of the PAIR model, the results are used to address the question of what minimum size asteroid constitutes a threat to the population. To answer this question, complete distributions of results are combined with a hypothetical risk tolerance posture to provide the minimum size, given sets of initial assumptions. Model outputs demonstrate how such questions can be answered and provide a means for interpreting the effect that input assumptions and uncertainty can have on final risk-based decisions. Model results can be used to prioritize investments to gain knowledge in critical areas or, conversely, to identify areas where additional data has little effect on the metrics of interest.

Reliability and cost evaluation of small isolated power systems containing photovoltaic and wind energy

NASA Astrophysics Data System (ADS)

Karki, Rajesh

Renewable energy application in electric power systems is growing rapidly worldwide due to enhanced public concerns for adverse environmental impacts and escalation in energy costs associated with the use of conventional energy sources. Photovoltaics and wind energy sources are being increasingly recognized as cost effective generation sources. A comprehensive evaluation of reliability and cost is required to analyze the actual benefits of utilizing these energy sources. The reliability aspects of utilizing renewable energy sources have largely been ignored in the past due the relatively insignificant contribution of these sources in major power systems, and consequently due to the lack of appropriate techniques. Renewable energy sources have the potential to play a significant role in the electrical energy requirements of small isolated power systems which are primarily supplied by costly diesel fuel. A relatively high renewable energy penetration can significantly reduce the system fuel costs but can also have considerable impact on the system reliability. Small isolated systems routinely plan their generating facilities using deterministic adequacy methods that cannot incorporate the highly erratic behavior of renewable energy sources. The utilization of a single probabilistic risk index has not been generally accepted in small isolated system evaluation despite its utilization in most large power utilities. Deterministic and probabilistic techniques are combined in this thesis using a system well-being approach to provide useful adequacy indices for small isolated systems that include renewable energy. This thesis presents an evaluation model for small isolated systems containing renewable energy sources by integrating simulation models that generate appropriate atmospheric data, evaluate chronological renewable power outputs and combine total available energy and load to provide useful system indices. A software tool SIPSREL+ has been developed which generates risk, well-being and energy based indices to provide realistic cost/reliability measures of utilizing renewable energy. The concepts presented and the examples illustrated in this thesis will help system planners to decide on appropriate installation sites, the types and mix of different energy generating sources, the optimum operating policies, and the optimum generation expansion plans required to meet increasing load demands in small isolated power systems containing photovoltaic and wind energy sources.

Charles Bonnet Syndrome: Evidence for a Generative Model in the Cortex?

PubMed Central

Reichert, David P.; Seriès, Peggy; Storkey, Amos J.

2013-01-01

Several theories propose that the cortex implements an internal model to explain, predict, and learn about sensory data, but the nature of this model is unclear. One condition that could be highly informative here is Charles Bonnet syndrome (CBS), where loss of vision leads to complex, vivid visual hallucinations of objects, people, and whole scenes. CBS could be taken as indication that there is a generative model in the brain, specifically one that can synthesise rich, consistent visual representations even in the absence of actual visual input. The processes that lead to CBS are poorly understood. Here, we argue that a model recently introduced in machine learning, the deep Boltzmann machine (DBM), could capture the relevant aspects of (hypothetical) generative processing in the cortex. The DBM carries both the semantics of a probabilistic generative model and of a neural network. The latter allows us to model a concrete neural mechanism that could underlie CBS, namely, homeostatic regulation of neuronal activity. We show that homeostatic plasticity could serve to make the learnt internal model robust against e.g. degradation of sensory input, but overcompensate in the case of CBS, leading to hallucinations. We demonstrate how a wide range of features of CBS can be explained in the model and suggest a potential role for the neuromodulator acetylcholine. This work constitutes the first concrete computational model of CBS and the first application of the DBM as a model in computational neuroscience. Our results lend further credence to the hypothesis of a generative model in the brain. PMID:23874177

Design optimization and probabilistic analysis of a hydrodynamic journal bearing

NASA Technical Reports Server (NTRS)

Liniecki, Alexander G.

1990-01-01

A nonlinear constrained optimization of a hydrodynamic bearing was performed yielding three main variables: radial clearance, bearing length to diameter ratio, and lubricating oil viscosity. As an objective function a combined model of temperature rise and oil supply has been adopted. The optimized model of the bearing has been simulated for population of 1000 cases using Monte Carlo statistical method. It appeared that the so called 'optimal solution' generated more than 50 percent of failed bearings, because their minimum oil film thickness violated stipulated minimum constraint value. As a remedy change of oil viscosity is suggested after several sensitivities of variables have been investigated.

Application of the LEPS technique for Quantitative Precipitation Forecasting (QPF) in Southern Italy: a preliminary study

NASA Astrophysics Data System (ADS)

Federico, S.; Avolio, E.; Bellecci, C.; Colacino, M.; Walko, R. L.

2006-03-01

This paper reports preliminary results for a Limited area model Ensemble Prediction System (LEPS), based on RAMS (Regional Atmospheric Modelling System), for eight case studies of moderate-intense precipitation over Calabria, the southernmost tip of the Italian peninsula. LEPS aims to transfer the benefits of a probabilistic forecast from global to regional scales in countries where local orographic forcing is a key factor to force convection. To accomplish this task and to limit computational time in an operational implementation of LEPS, we perform a cluster analysis of ECMWF-EPS runs. Starting from the 51 members that form the ECMWF-EPS we generate five clusters. For each cluster a representative member is selected and used to provide initial and dynamic boundary conditions to RAMS, whose integrations generate LEPS. RAMS runs have 12-km horizontal resolution. To analyze the impact of enhanced horizontal resolution on quantitative precipitation forecasts, LEPS forecasts are compared to a full Brute Force (BF) ensemble. This ensemble is based on RAMS, has 36 km horizontal resolution and is generated by 51 members, nested in each ECMWF-EPS member. LEPS and BF results are compared subjectively and by objective scores. Subjective analysis is based on precipitation and probability maps of case studies whereas objective analysis is made by deterministic and probabilistic scores. Scores and maps are calculated by comparing ensemble precipitation forecasts against reports from the Calabria regional raingauge network. Results show that LEPS provided better rainfall predictions than BF for all case studies selected. This strongly suggests the importance of the enhanced horizontal resolution, compared to ensemble population, for Calabria for these cases. To further explore the impact of local physiographic features on QPF (Quantitative Precipitation Forecasting), LEPS results are also compared with a 6-km horizontal resolution deterministic forecast. Due to local and mesoscale forcing, the high resolution forecast (Hi-Res) has better performance compared to the ensemble mean for rainfall thresholds larger than 10mm but it tends to overestimate precipitation for lower amounts. This yields larger false alarms that have a detrimental effect on objective scores for lower thresholds. To exploit the advantages of a probabilistic forecast compared to a deterministic one, the relation between the ECMWF-EPS 700 hPa geopotential height spread and LEPS performance is analyzed. Results are promising even if additional studies are required.

Constructing probabilistic scenarios for wide-area solar power generation

DOE PAGES

Woodruff, David L.; Deride, Julio; Staid, Andrea; ...

2017-12-22

Optimizing thermal generation commitments and dispatch in the presence of high penetrations of renewable resources such as solar energy requires a characterization of their stochastic properties. In this study, we describe novel methods designed to create day-ahead, wide-area probabilistic solar power scenarios based only on historical forecasts and associated observations of solar power production. Each scenario represents a possible trajectory for solar power in next-day operations with an associated probability computed by algorithms that use historical forecast errors. Scenarios are created by segmentation of historic data, fitting non-parametric error distributions using epi-splines, and then computing specific quantiles from these distributions.more » Additionally, we address the challenge of establishing an upper bound on solar power output. Our specific application driver is for use in stochastic variants of core power systems operations optimization problems, e.g., unit commitment and economic dispatch. These problems require as input a range of possible future realizations of renewables production. However, the utility of such probabilistic scenarios extends to other contexts, e.g., operator and trader situational awareness. Finally, we compare the performance of our approach to a recently proposed method based on quantile regression, and demonstrate that our method performs comparably to this approach in terms of two widely used methods for assessing the quality of probabilistic scenarios: the Energy score and the Variogram score.« less

Constructing probabilistic scenarios for wide-area solar power generation

DOE Office of Scientific and Technical Information (OSTI.GOV)

Woodruff, David L.; Deride, Julio; Staid, Andrea

Optimizing thermal generation commitments and dispatch in the presence of high penetrations of renewable resources such as solar energy requires a characterization of their stochastic properties. In this study, we describe novel methods designed to create day-ahead, wide-area probabilistic solar power scenarios based only on historical forecasts and associated observations of solar power production. Each scenario represents a possible trajectory for solar power in next-day operations with an associated probability computed by algorithms that use historical forecast errors. Scenarios are created by segmentation of historic data, fitting non-parametric error distributions using epi-splines, and then computing specific quantiles from these distributions.more » Additionally, we address the challenge of establishing an upper bound on solar power output. Our specific application driver is for use in stochastic variants of core power systems operations optimization problems, e.g., unit commitment and economic dispatch. These problems require as input a range of possible future realizations of renewables production. However, the utility of such probabilistic scenarios extends to other contexts, e.g., operator and trader situational awareness. Finally, we compare the performance of our approach to a recently proposed method based on quantile regression, and demonstrate that our method performs comparably to this approach in terms of two widely used methods for assessing the quality of probabilistic scenarios: the Energy score and the Variogram score.« less

Probabilistic Structural Analysis Methods (PSAM) for Select Space Propulsion System Components

NASA Technical Reports Server (NTRS)

1999-01-01

Probabilistic Structural Analysis Methods (PSAM) are described for the probabilistic structural analysis of engine components for current and future space propulsion systems. Components for these systems are subjected to stochastic thermomechanical launch loads. Uncertainties or randomness also occurs in material properties, structural geometry, and boundary conditions. Material property stochasticity, such as in modulus of elasticity or yield strength, exists in every structure and is a consequence of variations in material composition and manufacturing processes. Procedures are outlined for computing the probabilistic structural response or reliability of the structural components. The response variables include static or dynamic deflections, strains, and stresses at one or several locations, natural frequencies, fatigue or creep life, etc. Sample cases illustrates how the PSAM methods and codes simulate input uncertainties and compute probabilistic response or reliability using a finite element model with probabilistic methods.

Will current probabilistic climate change information, as such, improve adaptation?

NASA Astrophysics Data System (ADS)

Lopez, A.; Smith, L. A.

2012-04-01

Probabilistic climate scenarios are currently being provided to end users, to employ as probabilities in adaptation decision making, with the explicit suggestion that they quantify the impacts of climate change relevant to a variety of sectors. These "probabilities" are, however, rather sensitive to the assumptions in, and the structure of the modelling approaches used to generate them. It is often argued that stakeholders require probabilistic climate change information to adequately evaluate and plan adaptation pathways. On the other hand, some circumstantial evidence suggests that on the ground decision making rarely uses well defined probability distributions of climate change as inputs. Nevertheless it is within this context of probability distributions of climate change that we discuss possible drawbacks of supplying information that, while presented as robust and decision relevant, , is in fact unlikely to be so due to known flaws both in the underlying models and in the methodology used to "account for" those known flaws. How might one use a probability forecast that is expected to change in the future, not due to a refinement in our information but due to fundamental flaws in its construction? What then are the alternatives? While the answer will depend on the context of the problem at hand, a good approach will be strongly informed by the timescale of the given planning decision, and the consideration of all the non-climatic factors that have to be taken into account in the corresponding risk assessment. Using a water resources system as an example, we illustrate an alternative approach to deal with these challenges and make robust adaptation decisions today.

BAYESIAN PROTEIN STRUCTURE ALIGNMENT.

PubMed

Rodriguez, Abel; Schmidler, Scott C

The analysis of the three-dimensional structure of proteins is an important topic in molecular biochemistry. Structure plays a critical role in defining the function of proteins and is more strongly conserved than amino acid sequence over evolutionary timescales. A key challenge is the identification and evaluation of structural similarity between proteins; such analysis can aid in understanding the role of newly discovered proteins and help elucidate evolutionary relationships between organisms. Computational biologists have developed many clever algorithmic techniques for comparing protein structures, however, all are based on heuristic optimization criteria, making statistical interpretation somewhat difficult. Here we present a fully probabilistic framework for pairwise structural alignment of proteins. Our approach has several advantages, including the ability to capture alignment uncertainty and to estimate key "gap" parameters which critically affect the quality of the alignment. We show that several existing alignment methods arise as maximum a posteriori estimates under specific choices of prior distributions and error models. Our probabilistic framework is also easily extended to incorporate additional information, which we demonstrate by including primary sequence information to generate simultaneous sequence-structure alignments that can resolve ambiguities obtained using structure alone. This combined model also provides a natural approach for the difficult task of estimating evolutionary distance based on structural alignments. The model is illustrated by comparison with well-established methods on several challenging protein alignment examples.

Estimation of distribution algorithm with path relinking for the blocking flow-shop scheduling problem

NASA Astrophysics Data System (ADS)

Shao, Zhongshi; Pi, Dechang; Shao, Weishi

2018-05-01

This article presents an effective estimation of distribution algorithm, named P-EDA, to solve the blocking flow-shop scheduling problem (BFSP) with the makespan criterion. In the P-EDA, a Nawaz-Enscore-Ham (NEH)-based heuristic and the random method are combined to generate the initial population. Based on several superior individuals provided by a modified linear rank selection, a probabilistic model is constructed to describe the probabilistic distribution of the promising solution space. The path relinking technique is incorporated into EDA to avoid blindness of the search and improve the convergence property. A modified referenced local search is designed to enhance the local exploitation. Moreover, a diversity-maintaining scheme is introduced into EDA to avoid deterioration of the population. Finally, the parameters of the proposed P-EDA are calibrated using a design of experiments approach. Simulation results and comparisons with some well-performing algorithms demonstrate the effectiveness of the P-EDA for solving BFSP.

A Bayesian-based two-stage inexact optimization method for supporting stream water quality management in the Three Gorges Reservoir region.

PubMed

Hu, X H; Li, Y P; Huang, G H; Zhuang, X W; Ding, X W

2016-05-01

In this study, a Bayesian-based two-stage inexact optimization (BTIO) method is developed for supporting water quality management through coupling Bayesian analysis with interval two-stage stochastic programming (ITSP). The BTIO method is capable of addressing uncertainties caused by insufficient inputs in water quality model as well as uncertainties expressed as probabilistic distributions and interval numbers. The BTIO method is applied to a real case of water quality management for the Xiangxi River basin in the Three Gorges Reservoir region to seek optimal water quality management schemes under various uncertainties. Interval solutions for production patterns under a range of probabilistic water quality constraints have been generated. Results obtained demonstrate compromises between the system benefit and the system failure risk due to inherent uncertainties that exist in various system components. Moreover, information about pollutant emission is accomplished, which would help managers to adjust production patterns of regional industry and local policies considering interactions of water quality requirement, economic benefit, and industry structure.

Simulation investigation of multipactor in metal components for space application with an improved secondary emission model

DOE Office of Scientific and Technical Information (OSTI.GOV)

Li, Yun, E-mail: genliyun@126.com, E-mail: cuiwanzhao@126.com; Cui, Wan-Zhao, E-mail: genliyun@126.com, E-mail: cuiwanzhao@126.com; Wang, Hong-Guang

2015-05-15

Effects of the secondary electron emission (SEE) phenomenon of metal surface on the multipactor analysis of microwave components are investigated numerically and experimentally in this paper. Both the secondary electron yield (SEY) and the emitted energy spectrum measurements are performed on silver plated samples for accurate description of the SEE phenomenon. A phenomenological probabilistic model based on SEE physics is utilized and fitted accurately to the measured SEY and emitted energy spectrum of the conditioned surface material of microwave components. Specially, the phenomenological probabilistic model is extended to the low primary energy end lower than 20 eV mathematically, since no accuratemore » measurement data can be obtained. Embedding the phenomenological probabilistic model into the Electromagnetic Particle-In-Cell (EM-PIC) method, the electronic resonant multipacting in microwave components can be tracked and hence the multipactor threshold can be predicted. The threshold prediction error of the transformer and the coaxial filter is 0.12 dB and 1.5 dB, respectively. Simulation results demonstrate that the discharge threshold is strongly dependent on the SEYs and its energy spectrum in the low energy end (lower than 50 eV). Multipacting simulation results agree quite well with experiments in practical components, while the phenomenological probabilistic model fit both the SEY and the emission energy spectrum better than the traditionally used model and distribution. The EM-PIC simulation method with the phenomenological probabilistic model for the surface collision simulation has been demonstrated for predicting the multipactor threshold in metal components for space application.« less

Genovo: De Novo Assembly for Metagenomes

NASA Astrophysics Data System (ADS)

Laserson, Jonathan; Jojic, Vladimir; Koller, Daphne

Next-generation sequencing technologies produce a large number of noisy reads from the DNA in a sample. Metagenomics and population sequencing aim to recover the genomic sequences of the species in the sample, which could be of high diversity. Methods geared towards single sequence reconstruction are not sensitive enough when applied in this setting. We introduce a generative probabilistic model of read generation from environmental samples and present Genovo, a novel de novo sequence assembler that discovers likely sequence reconstructions under the model. A Chinese restaurant process prior accounts for the unknown number of genomes in the sample. Inference is made by applying a series of hill-climbing steps iteratively until convergence. We compare the performance of Genovo to three other short read assembly programs across one synthetic dataset and eight metagenomic datasets created using the 454 platform, the largest of which has 311k reads. Genovo's reconstructions cover more bases and recover more genes than the other methods, and yield a higher assembly score.

A transient stochastic weather generator incorporating climate model uncertainty

NASA Astrophysics Data System (ADS)

Glenis, Vassilis; Pinamonti, Valentina; Hall, Jim W.; Kilsby, Chris G.

2015-11-01

Stochastic weather generators (WGs), which provide long synthetic time series of weather variables such as rainfall and potential evapotranspiration (PET), have found widespread use in water resources modelling. When conditioned upon the changes in climatic statistics (change factors, CFs) predicted by climate models, WGs provide a useful tool for climate impacts assessment and adaption planning. The latest climate modelling exercises have involved large numbers of global and regional climate models integrations, designed to explore the implications of uncertainties in the climate model formulation and parameter settings: so called 'perturbed physics ensembles' (PPEs). In this paper we show how these climate model uncertainties can be propagated through to impact studies by testing multiple vectors of CFs, each vector derived from a different sample from a PPE. We combine this with a new methodology to parameterise the projected time-evolution of CFs. We demonstrate how, when conditioned upon these time-dependent CFs, an existing, well validated and widely used WG can be used to generate non-stationary simulations of future climate that are consistent with probabilistic outputs from the Met Office Hadley Centre's Perturbed Physics Ensemble. The WG enables extensive sampling of natural variability and climate model uncertainty, providing the basis for development of robust water resources management strategies in the context of a non-stationary climate.

FAVOR: A new fracture mechanics code for reactor pressure vessels subjected to pressurized thermal shock

DOE Office of Scientific and Technical Information (OSTI.GOV)

Dickson, T.L.

1993-01-01

This report discusses probabilistic fracture mechanics (PFM) analysis which is a major element of the comprehensive probabilistic methodology endorsed by the NRC for evaluation of the integrity of Pressurized Water Reactor (PWR) pressure vessels subjected to pressurized-thermal-shock (PTS) transients. It is anticipated that there will be an increasing need for an improved and validated PTS PFM code which is accepted by the NRC and utilities, as more plants approach the PTS screening criteria and are required to perform plant-specific analyses. The NRC funded Heavy Section Steel Technology (HSST) Program at Oak Ridge National Laboratories is currently developing the FAVOR (Fracturemore » Analysis of Vessels: Oak Ridge) PTS PFM code, which is intended to meet this need. The FAVOR code incorporates the most important features of both OCA-P and VISA-II and contains some new capabilities such as PFM global modeling methodology, the capability to approximate the effects of thermal streaming on circumferential flaws located inside a plume region created by fluid and thermal stratification, a library of stress intensity factor influence coefficients, generated by the NQA-1 certified ABAQUS computer code, for an adequate range of two and three dimensional inside surface flaws, the flexibility to generate a variety of output reports, and user friendliness.« less

FAVOR: A new fracture mechanics code for reactor pressure vessels subjected to pressurized thermal shock

DOE Office of Scientific and Technical Information (OSTI.GOV)

Dickson, T.L.

1993-04-01

This report discusses probabilistic fracture mechanics (PFM) analysis which is a major element of the comprehensive probabilistic methodology endorsed by the NRC for evaluation of the integrity of Pressurized Water Reactor (PWR) pressure vessels subjected to pressurized-thermal-shock (PTS) transients. It is anticipated that there will be an increasing need for an improved and validated PTS PFM code which is accepted by the NRC and utilities, as more plants approach the PTS screening criteria and are required to perform plant-specific analyses. The NRC funded Heavy Section Steel Technology (HSST) Program at Oak Ridge National Laboratories is currently developing the FAVOR (Fracturemore » Analysis of Vessels: Oak Ridge) PTS PFM code, which is intended to meet this need. The FAVOR code incorporates the most important features of both OCA-P and VISA-II and contains some new capabilities such as PFM global modeling methodology, the capability to approximate the effects of thermal streaming on circumferential flaws located inside a plume region created by fluid and thermal stratification, a library of stress intensity factor influence coefficients, generated by the NQA-1 certified ABAQUS computer code, for an adequate range of two and three dimensional inside surface flaws, the flexibility to generate a variety of output reports, and user friendliness.« less

Improving real-time inflow forecasting into hydropower reservoirs through a complementary modelling framework

NASA Astrophysics Data System (ADS)

Gragne, A. S.; Sharma, A.; Mehrotra, R.; Alfredsen, K.

2015-08-01

Accuracy of reservoir inflow forecasts is instrumental for maximizing the value of water resources and benefits gained through hydropower generation. Improving hourly reservoir inflow forecasts over a 24 h lead time is considered within the day-ahead (Elspot) market of the Nordic exchange market. A complementary modelling framework presents an approach for improving real-time forecasting without needing to modify the pre-existing forecasting model, but instead formulating an independent additive or complementary model that captures the structure the existing operational model may be missing. We present here the application of this principle for issuing improved hourly inflow forecasts into hydropower reservoirs over extended lead times, and the parameter estimation procedure reformulated to deal with bias, persistence and heteroscedasticity. The procedure presented comprises an error model added on top of an unalterable constant parameter conceptual model. This procedure is applied in the 207 km2 Krinsvatn catchment in central Norway. The structure of the error model is established based on attributes of the residual time series from the conceptual model. Besides improving forecast skills of operational models, the approach estimates the uncertainty in the complementary model structure and produces probabilistic inflow forecasts that entrain suitable information for reducing uncertainty in the decision-making processes in hydropower systems operation. Deterministic and probabilistic evaluations revealed an overall significant improvement in forecast accuracy for lead times up to 17 h. Evaluation of the percentage of observations bracketed in the forecasted 95 % confidence interval indicated that the degree of success in containing 95 % of the observations varies across seasons and hydrologic years.

Probabilistic choice models in health-state valuation research: background, theories, assumptions and applications.

PubMed

Arons, Alexander M M; Krabbe, Paul F M

2013-02-01

Interest is rising in measuring subjective health outcomes, such as treatment outcomes that are not directly quantifiable (functional disability, symptoms, complaints, side effects and health-related quality of life). Health economists in particular have applied probabilistic choice models in the area of health evaluation. They increasingly use discrete choice models based on random utility theory to derive values for healthcare goods or services. Recent attempts have been made to use discrete choice models as an alternative method to derive values for health states. In this article, various probabilistic choice models are described according to their underlying theory. A historical overview traces their development and applications in diverse fields. The discussion highlights some theoretical and technical aspects of the choice models and their similarity and dissimilarity. The objective of the article is to elucidate the position of each model and their applications for health-state valuation.

Inherent limitations of probabilistic models for protein-DNA binding specificity

PubMed Central

Ruan, Shuxiang

2017-01-01

The specificities of transcription factors are most commonly represented with probabilistic models. These models provide a probability for each base occurring at each position within the binding site and the positions are assumed to contribute independently. The model is simple and intuitive and is the basis for many motif discovery algorithms. However, the model also has inherent limitations that prevent it from accurately representing true binding probabilities, especially for the highest affinity sites under conditions of high protein concentration. The limitations are not due to the assumption of independence between positions but rather are caused by the non-linear relationship between binding affinity and binding probability and the fact that independent normalization at each position skews the site probabilities. Generally probabilistic models are reasonably good approximations, but new high-throughput methods allow for biophysical models with increased accuracy that should be used whenever possible. PMID:28686588

Bias Characterization in Probabilistic Genotype Data and Improved Signal Detection with Multiple Imputation

PubMed Central

Palmer, Cameron; Pe’er, Itsik

2016-01-01

Missing data are an unavoidable component of modern statistical genetics. Different array or sequencing technologies cover different single nucleotide polymorphisms (SNPs), leading to a complicated mosaic pattern of missingness where both individual genotypes and entire SNPs are sporadically absent. Such missing data patterns cannot be ignored without introducing bias, yet cannot be inferred exclusively from nonmissing data. In genome-wide association studies, the accepted solution to missingness is to impute missing data using external reference haplotypes. The resulting probabilistic genotypes may be analyzed in the place of genotype calls. A general-purpose paradigm, called Multiple Imputation (MI), is known to model uncertainty in many contexts, yet it is not widely used in association studies. Here, we undertake a systematic evaluation of existing imputed data analysis methods and MI. We characterize biases related to uncertainty in association studies, and find that bias is introduced both at the imputation level, when imputation algorithms generate inconsistent genotype probabilities, and at the association level, when analysis methods inadequately model genotype uncertainty. We find that MI performs at least as well as existing methods or in some cases much better, and provides a straightforward paradigm for adapting existing genotype association methods to uncertain data. PMID:27310603

Probabilistic modelling of flood events using the entropy copula

NASA Astrophysics Data System (ADS)

Li, Fan; Zheng, Qian

2016-11-01

The estimation of flood frequency is vital for the flood control strategies and hydraulic structure design. Generating synthetic flood events according to statistical properties of observations is one of plausible methods to analyze the flood frequency. Due to the statistical dependence among the flood event variables (i.e. the flood peak, volume and duration), a multidimensional joint probability estimation is required. Recently, the copula method is widely used for multivariable dependent structure construction, however, the copula family should be chosen before application and the choice process is sometimes rather subjective. The entropy copula, a new copula family, employed in this research proposed a way to avoid the relatively subjective process by combining the theories of copula and entropy. The analysis shows the effectiveness of the entropy copula for probabilistic modelling the flood events of two hydrological gauges, and a comparison of accuracy with the popular copulas was made. The Gibbs sampling technique was applied for trivariate flood events simulation in order to mitigate the calculation difficulties of extending to three dimension directly. The simulation results indicate that the entropy copula is a simple and effective copula family for trivariate flood simulation.

Probabilistic HR Diagrams: A New Infrared and X-ray Chronometer for Very Young, Massive Stellar Clusters and Associations

NASA Astrophysics Data System (ADS)

Maldonado, Jessica; Povich, Matthew S.

2016-01-01

We present a novel method for constraining the duration of star formation in very young, massive star-forming regions. Constraints on stellar population ages are derived from probabilistic HR diagrams (pHRDs) generated by fitting stellar model spectra to the infrared (IR) spectral energy distributions (SEDs) of Herbig Ae/Be stars and their less-evolved, pre-main sequence progenitors. Stellar samples for the pHRDs are selected based on the detection of X-ray emission associated with the IR source, and the lack of detectible IR excess emission at wavelengths ≤4.5 µm. The SED model fits were used to create two-dimensional probability distributions of the stellar parameters, specifically bolometric luminosity versus temperature and mass versus evolutionary age. We present first results from the pHRD analysis of the relatively evolved Carina Nebula and the unevolved M17 SWex infrared dark cloud, which reveal the expected, strikingly different star formation durations between these two regions. In the future, we will apply this method to analyze available X-ray and IR data from the MYStIX project on other Galactic massive star forming regions within 3 kpc of the Sun.

Capturing spatial and temporal patterns of widespread, extreme flooding across Europe

NASA Astrophysics Data System (ADS)

Busby, Kathryn; Raven, Emma; Liu, Ye

2013-04-01

Statistical characterisation of physical hazards is an integral part of probabilistic catastrophe models used by the reinsurance industry to estimate losses from large scale events. Extreme flood events are not restricted by country boundaries which poses an issue for reinsurance companies as their exposures often extend beyond them. We discuss challenges and solutions that allow us to appropriately capture the spatial and temporal dependence of extreme hydrological events on a continental-scale, which in turn enables us to generate an industry-standard stochastic event set for estimating financial losses for widespread flooding. By presenting our event set methodology, we focus on explaining how extreme value theory (EVT) and dependence modelling are used to account for short, inconsistent hydrological data from different countries, and how to make appropriate statistical decisions that best characterise the nature of flooding across Europe. The consistency of input data is of vital importance when identifying historical flood patterns. Collating data from numerous sources inherently causes inconsistencies and we demonstrate our robust approach to assessing the data and refining it to compile a single consistent dataset. This dataset is then extrapolated using a parameterised EVT distribution to estimate extremes. Our method then captures the dependence of flood events across countries using an advanced multivariate extreme value model. Throughout, important statistical decisions are explored including: (1) distribution choice; (2) the threshold to apply for extracting extreme data points; (3) a regional analysis; (4) the definition of a flood event, which is often linked with reinsurance industry's hour's clause; and (5) handling of missing values. Finally, having modelled the historical patterns of flooding across Europe, we sample from this model to generate our stochastic event set comprising of thousands of events over thousands of years. We then briefly illustrate how this is applied within a probabilistic model to estimate catastrophic loss curves used by the reinsurance industry.

A Probabilistic Tsunami Hazard Assessment Methodology and Its Application to Crescent City, CA

NASA Astrophysics Data System (ADS)

Gonzalez, F. I.; Leveque, R. J.; Waagan, K.; Adams, L.; Lin, G.

2012-12-01

A PTHA methodology, based in large part on Probabilistic Seismic Hazard Assessment methods (e.g., Cornell, 1968; SSHAC, 1997; Geist and Parsons, 2005), was previously applied to Seaside, OR (Gonzalez, et al., 2009). This initial version of the method has been updated to include: a revised method to estimate tidal uncertainty; an improved method for generating stochastic realizations to estimate slip distribution uncertainty (Mai and Beroza, 2002; Blair, et al., 2011); additional near-field sources in the Cascadia Subduction Zone, based on the work of Goldfinger, et al. (2012); far-field sources in Japan, based on information updated since the 3 March 2011 Tohoku tsunami (Japan Earthquake Research Committee, 2011). The GeoClaw tsunami model (Berger, et. al, 2011) is used to simulate generation, propagation and inundation. We will discuss this revised PTHA methodology and the results of its application to Crescent City, CA. Berger, M.J., D. L. George, R. J. LeVeque, and K. T. Mandli, The GeoClaw software for depth-averaged flows with adaptive refinement, Adv. Water Res. 34 (2011), pp. 1195-1206. Blair, J.L., McCrory, P.A., Oppenheimer, D.H., and Waldhauser, F. (2011): A Geo-referenced 3D model of the Juan de Fuca Slab and associated seismicity: U.S. Geological Survey Data Series 633, v.1.0, available at http://pubs.usgs.gov/ds/633/. Cornell, C. A. (1968): Engineering seismic risk analysis, Bull. Seismol. Soc. Am., 58, 1583-1606. Geist, E. L., and T. Parsons (2005): Probabilistic Analysis of Tsunami Hazards, Nat. Hazards, 37 (3), 277-314. Goldfinger, C., Nelson, C.H., Morey, A.E., Johnson, J.E., Patton, J.R., Karabanov, E., Gutiérrez-Pastor, J., Eriksson, A.T., Gràcia, E., Dunhill, G., Enkin, R.J., Dallimore, A., and Vallier, T. (2012): Turbidite event history—Methods and implications for Holocene paleoseismicity of the Cascadia subduction zone: U.S. Geological Survey Professional Paper 1661-F, 170 p. (Available at http://pubs.usgs.gov/pp/pp1661f/). González, F.I., E.L. Geist, B. Jaffe, U. Kânoglu, H. Mofjeld, C.E. Synolakis, V.V Titov, D. Arcas, D. Bellomo, D. Carlton, T. Horning, J. Johnson, J. Newman, T. Parsons, R. Peters, C. Peterson, G .Priest, A. Venturato, J. Weber, F. Wong, and A. Yalciner (2009): Probabilistic Tsunami Hazard Assessment at Seaside, Oregon, for Near- and Far-Field Seismic Sources, J. Geophys. Res., 114, C11023, doi:10.1029/2008JC005132. Japan Earthquake Research Committee, (2011): http://www.jishin.go.jp/main/p_hyoka02.htm Mai, P. M., and G. C. Beroza (2002): A spatial random field model to characterize complexity in earthquake slip, J. Geophys. Res., 107(B11), 2308, doi:10.1029/2001JB000588. SSHAC (Senior Seismic Hazard Analysis Committee) (1997): Recommendations for Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of Experts, Main Report Rep. NUREG/CR-6372 UCRL-ID-122160 Vol. 1, 256 pp, U.S. Nuclear Regulatory Commission.

A probabilistic multi-criteria decision making technique for conceptual and preliminary aerospace systems design

NASA Astrophysics Data System (ADS)

Bandte, Oliver

It has always been the intention of systems engineering to invent or produce the best product possible. Many design techniques have been introduced over the course of decades that try to fulfill this intention. Unfortunately, no technique has succeeded in combining multi-criteria decision making with probabilistic design. The design technique developed in this thesis, the Joint Probabilistic Decision Making (JPDM) technique, successfully overcomes this deficiency by generating a multivariate probability distribution that serves in conjunction with a criterion value range of interest as a universally applicable objective function for multi-criteria optimization and product selection. This new objective function constitutes a meaningful Xnetric, called Probability of Success (POS), that allows the customer or designer to make a decision based on the chance of satisfying the customer's goals. In order to incorporate a joint probabilistic formulation into the systems design process, two algorithms are created that allow for an easy implementation into a numerical design framework: the (multivariate) Empirical Distribution Function and the Joint Probability Model. The Empirical Distribution Function estimates the probability that an event occurred by counting how many times it occurred in a given sample. The Joint Probability Model on the other hand is an analytical parametric model for the multivariate joint probability. It is comprised of the product of the univariate criterion distributions, generated by the traditional probabilistic design process, multiplied with a correlation function that is based on available correlation information between pairs of random variables. JPDM is an excellent tool for multi-objective optimization and product selection, because of its ability to transform disparate objectives into a single figure of merit, the likelihood of successfully meeting all goals or POS. The advantage of JPDM over other multi-criteria decision making techniques is that POS constitutes a single optimizable function or metric that enables a comparison of all alternative solutions on an equal basis. Hence, POS allows for the use of any standard single-objective optimization technique available and simplifies a complex multi-criteria selection problem into a simple ordering problem, where the solution with the highest POS is best. By distinguishing between controllable and uncontrollable variables in the design process, JPDM can account for the uncertain values of the uncontrollable variables that are inherent to the design problem, while facilitating an easy adjustment of the controllable ones to achieve the highest possible POS. Finally, JPDM's superiority over current multi-criteria decision making techniques is demonstrated with an optimization of a supersonic transport concept and ten contrived equations as well as a product selection example, determining an airline's best choice among Boeing's B-747, B-777, Airbus' A340, and a Supersonic Transport. The optimization examples demonstrate JPDM's ability to produce a better solution with a higher POS than an Overall Evaluation Criterion or Goal Programming approach. Similarly, the product selection example demonstrates JPDM's ability to produce a better solution with a higher POS and different ranking than the Overall Evaluation Criterion or Technique for Order Preferences by Similarity to the Ideal Solution (TOPSIS) approach.

Probabilistic Learning by Rodent Grid Cells

PubMed Central

Cheung, Allen

2016-01-01

Mounting evidence shows mammalian brains are probabilistic computers, but the specific cells involved remain elusive. Parallel research suggests that grid cells of the mammalian hippocampal formation are fundamental to spatial cognition but their diverse response properties still defy explanation. No plausible model exists which explains stable grids in darkness for twenty minutes or longer, despite being one of the first results ever published on grid cells. Similarly, no current explanation can tie together grid fragmentation and grid rescaling, which show very different forms of flexibility in grid responses when the environment is varied. Other properties such as attractor dynamics and grid anisotropy seem to be at odds with one another unless additional properties are assumed such as a varying velocity gain. Modelling efforts have largely ignored the breadth of response patterns, while also failing to account for the disastrous effects of sensory noise during spatial learning and recall, especially in darkness. Here, published electrophysiological evidence from a range of experiments are reinterpreted using a novel probabilistic learning model, which shows that grid cell responses are accurately predicted by a probabilistic learning process. Diverse response properties of probabilistic grid cells are statistically indistinguishable from rat grid cells across key manipulations. A simple coherent set of probabilistic computations explains stable grid fields in darkness, partial grid rescaling in resized arenas, low-dimensional attractor grid cell dynamics, and grid fragmentation in hairpin mazes. The same computations also reconcile oscillatory dynamics at the single cell level with attractor dynamics at the cell ensemble level. Additionally, a clear functional role for boundary cells is proposed for spatial learning. These findings provide a parsimonious and unified explanation of grid cell function, and implicate grid cells as an accessible neuronal population readout of a set of probabilistic spatial computations. PMID:27792723

A Probabilistic Model for Diagnosing Misconceptions by a Pattern Classification Approach.

ERIC Educational Resources Information Center

Tatsuoka, Kikumi K.

A probabilistic approach is introduced to classify and diagnose erroneous rules of operation resulting from a variety of misconceptions ("bugs") in a procedural domain of arithmetic. The model is contrasted with the deterministic approach which has commonly been used in the field of artificial intelligence, and the advantage of treating the…

Nonlinear probabilistic finite element models of laminated composite shells

NASA Technical Reports Server (NTRS)

Engelstad, S. P.; Reddy, J. N.

1993-01-01

A probabilistic finite element analysis procedure for laminated composite shells has been developed. A total Lagrangian finite element formulation, employing a degenerated 3-D laminated composite shell with the full Green-Lagrange strains and first-order shear deformable kinematics, forms the modeling foundation. The first-order second-moment technique for probabilistic finite element analysis of random fields is employed and results are presented in the form of mean and variance of the structural response. The effects of material nonlinearity are included through the use of a rate-independent anisotropic plasticity formulation with the macroscopic point of view. Both ply-level and micromechanics-level random variables can be selected, the latter by means of the Aboudi micromechanics model. A number of sample problems are solved to verify the accuracy of the procedures developed and to quantify the variability of certain material type/structure combinations. Experimental data is compared in many cases, and the Monte Carlo simulation method is used to check the probabilistic results. In general, the procedure is quite effective in modeling the mean and variance response of the linear and nonlinear behavior of laminated composite shells.

A unified probabilistic approach to improve spelling in an event-related potential-based brain-computer interface.

PubMed

Kindermans, Pieter-Jan; Verschore, Hannes; Schrauwen, Benjamin

2013-10-01

In recent years, in an attempt to maximize performance, machine learning approaches for event-related potential (ERP) spelling have become more and more complex. In this paper, we have taken a step back as we wanted to improve the performance without building an overly complex model, that cannot be used by the community. Our research resulted in a unified probabilistic model for ERP spelling, which is based on only three assumptions and incorporates language information. On top of that, the probabilistic nature of our classifier yields a natural dynamic stopping strategy. Furthermore, our method uses the same parameters across 25 subjects from three different datasets. We show that our classifier, when enhanced with language models and dynamic stopping, improves the spelling speed and accuracy drastically. Additionally, we would like to point out that as our model is entirely probabilistic, it can easily be used as the foundation for complex systems in future work. All our experiments are executed on publicly available datasets to allow for future comparison with similar techniques.

Reliability and cost/worth evaluation of generating systems utilizing wind and solar energy

NASA Astrophysics Data System (ADS)

Bagen

The utilization of renewable energy resources such as wind and solar energy for electric power supply has received considerable attention in recent years due to adverse environmental impacts and fuel cost escalation associated with conventional generation. At the present time, wind and/or solar energy sources are utilized to generate electric power in many applications. Wind and solar energy will become important sources for power generation in the future because of their environmental, social and economic benefits, together with public support and government incentives. The wind and sunlight are, however, unstable and variable energy sources, and behave far differently than conventional sources. Energy storage systems are, therefore, often required to smooth the fluctuating nature of the energy conversion system especially in small isolated applications. The research work presented in this thesis is focused on the development and application of reliability and economic benefits assessment associated with incorporating wind energy, solar energy and energy storage in power generating systems. A probabilistic approach using sequential Monte Carlo simulation was employed in this research and a number of analyses were conducted with regards to the adequacy and economic assessment of generation systems containing wind energy, solar energy and energy storage. The evaluation models and techniques incorporate risk index distributions and different operating strategies associated with diesel generation in small isolated systems. Deterministic and probabilistic techniques are combined in this thesis using a system well-being approach to provide useful adequacy indices for small isolated systems that include renewable energy and energy storage. The concepts presented and examples illustrated in this thesis will help power system planners and utility managers to assess the reliability and economic benefits of utilizing wind energy conversion systems, solar energy conversion systems and energy storage in electric power systems and provide useful input to the managerial decision process.

Probabilistic Inference in General Graphical Models through Sampling in Stochastic Networks of Spiking Neurons

PubMed Central

Pecevski, Dejan; Buesing, Lars; Maass, Wolfgang

2011-01-01

An important open problem of computational neuroscience is the generic organization of computations in networks of neurons in the brain. We show here through rigorous theoretical analysis that inherent stochastic features of spiking neurons, in combination with simple nonlinear computational operations in specific network motifs and dendritic arbors, enable networks of spiking neurons to carry out probabilistic inference through sampling in general graphical models. In particular, it enables them to carry out probabilistic inference in Bayesian networks with converging arrows (“explaining away”) and with undirected loops, that occur in many real-world tasks. Ubiquitous stochastic features of networks of spiking neurons, such as trial-to-trial variability and spontaneous activity, are necessary ingredients of the underlying computational organization. We demonstrate through computer simulations that this approach can be scaled up to neural emulations of probabilistic inference in fairly large graphical models, yielding some of the most complex computations that have been carried out so far in networks of spiking neurons. PMID:22219717

A Guide to the Literature on Learning Graphical Models

NASA Technical Reports Server (NTRS)

Buntine, Wray L.; Friedland, Peter (Technical Monitor)

1994-01-01

This literature review discusses different methods under the general rubric of learning Bayesian networks from data, and more generally, learning probabilistic graphical models. Because many problems in artificial intelligence, statistics and neural networks can be represented as a probabilistic graphical model, this area provides a unifying perspective on learning. This paper organizes the research in this area along methodological lines of increasing complexity.

Probabilistic parameter estimation of activated sludge processes using Markov Chain Monte Carlo.

PubMed

Sharifi, Soroosh; Murthy, Sudhir; Takács, Imre; Massoudieh, Arash

2014-03-01

One of the most important challenges in making activated sludge models (ASMs) applicable to design problems is identifying the values of its many stoichiometric and kinetic parameters. When wastewater characteristics data from full-scale biological treatment systems are used for parameter estimation, several sources of uncertainty, including uncertainty in measured data, external forcing (e.g. influent characteristics), and model structural errors influence the value of the estimated parameters. This paper presents a Bayesian hierarchical modeling framework for the probabilistic estimation of activated sludge process parameters. The method provides the joint probability density functions (JPDFs) of stoichiometric and kinetic parameters by updating prior information regarding the parameters obtained from expert knowledge and literature. The method also provides the posterior correlations between the parameters, as well as a measure of sensitivity of the different constituents with respect to the parameters. This information can be used to design experiments to provide higher information content regarding certain parameters. The method is illustrated using the ASM1 model to describe synthetically generated data from a hypothetical biological treatment system. The results indicate that data from full-scale systems can narrow down the ranges of some parameters substantially whereas the amount of information they provide regarding other parameters is small, due to either large correlations between some of the parameters or a lack of sensitivity with respect to the parameters. Copyright © 2013 Elsevier Ltd. All rights reserved.

Lessons learnt from tropical cyclone losses

NASA Astrophysics Data System (ADS)

Honegger, Caspar; Wüest, Marc; Zimmerli, Peter; Schoeck, Konrad

2016-04-01

Swiss Re has a long history in developing natural catastrophe loss models. The tropical cyclone USA and China model are examples for event-based models in their second generation. Both are based on basin-wide probabilistic track sets and calculate explicitly the losses from the sub-perils wind and storm surge in an insurance portfolio. Based on these models, we present two cases studies. China: a view on recent typhoon loss history Over the last 20 years only very few major tropical cyclones have caused severe insurance losses in the Pearl River Delta region and Shanghai, the two main exposure clusters along China's southeast coast. Several storms have made landfall in China every year but most struck areas with relatively low insured values. With this study, we make the point that typhoon landfalls in China have a strong hit-or-miss character and available insured loss experience is too short to form a representative view of risk. Historical storm tracks and a simple loss model applied to a market portfolio - all from publicly available data - are sufficient to illustrate this. An event-based probabilistic model is necessary for a reliable judgement of the typhoon risk in China. New York: current and future tropical cyclone risk In the aftermath of hurricane Sandy 2012, Swiss Re supported the City of New York in identifying ways to significantly improve the resilience to severe weather and climate change. Swiss Re provided a quantitative assessment of potential climate related risks facing the city as well as measures that could reduce those impacts.

Added value of non-calibrated and BMA calibrated AEMET-SREPS probabilistic forecasts: the 24 January 2009 extreme wind event over Catalonia

NASA Astrophysics Data System (ADS)

Escriba, P. A.; Callado, A.; Santos, D.; Santos, C.; Simarro, J.; García-Moya, J. A.

2009-09-01

At 00 UTC 24 January 2009 an explosive ciclogenesis originated over the Atlantic Ocean reached its maximum intensity with observed surface pressures lower than 970 hPa on its center and placed at Gulf of Vizcaya. During its path through southern France this low caused strong westerly and north-westerly winds over the Iberian Peninsula higher than 150 km/h at some places. These extreme winds leaved 10 casualties in Spain, 8 of them in Catalonia. The aim of this work is to show whether exists an added value in the short range prediction of the 24 January 2009 strong winds when using the Short Range Ensemble Prediction System (SREPS) of the Spanish Meteorological Agency (AEMET), with respect to the operational forecasting tools. This study emphasizes two aspects of probabilistic forecasting: the ability of a 3-day forecast of warn an extreme windy event and the ability of quantifying the predictability of the event so that giving value to deterministic forecast. Two type of probabilistic forecasts of wind are carried out, a non-calibrated and a calibrated one using Bayesian Model Averaging (BMA). AEMET runs daily experimentally SREPS twice a day (00 and 12 UTC). This system consists of 20 members that are constructed by integrating 5 local area models, COSMO (COSMO), HIRLAM (HIRLAM Consortium), HRM (DWD), MM5 (NOAA) and UM (UKMO), at 25 km of horizontal resolution. Each model uses 4 different initial and boundary conditions, the global models GFS (NCEP), GME (DWD), IFS (ECMWF) and UM. By this way it is obtained a probabilistic forecast that takes into account the initial, the contour and the model errors. BMA is a statistical tool for combining predictive probability functions from different sources. The BMA predictive probability density function (PDF) is a weighted average of PDFs centered on the individual bias-corrected forecasts. The weights are equal to posterior probabilities of the models generating the forecasts and reflect the skill of the ensemble members. Here BMA is applied to provide probabilistic forecasts of wind speed. In this work several forecasts for different time ranges (H+72, H+48 and H+24) of 10 meters wind speed over Catalonia are verified subjectively at one of the instants of maximum intensity, 12 UTC 24 January 2009. On one hand, three probabilistic forecasts are compared, ECMWF EPS, non-calibrated SREPS and calibrated SREPS. On the other hand, the relationship between predictability and skill of deterministic forecast is studied by looking at HIRLAM 0.16 deterministic forecasts of the event. Verification is focused on location and intensity of 10 meters wind speed and 10-minutal measures from AEMET automatic ground stations are used as observations. The results indicate that SREPS is able to forecast three days ahead mean winds higher than 36 km/h and that correctly localizes them with a significant probability of ocurrence in the affected area. The probability is higher after BMA calibration of the ensemble. The fact that probability of strong winds is high allows us to state that the predictability of the event is also high and, as a consequence, deterministic forecasts are more reliable. This is confirmed when verifying HIRLAM deterministic forecasts against observed values.

A probabilistic assessment of large scale wind power development for long-term energy resource planning

NASA Astrophysics Data System (ADS)

Kennedy, Scott Warren

A steady decline in the cost of wind turbines and increased experience in their successful operation have brought this technology to the forefront of viable alternatives for large-scale power generation. Methodologies for understanding the costs and benefits of large-scale wind power development, however, are currently limited. In this thesis, a new and widely applicable technique for estimating the social benefit of large-scale wind power production is presented. The social benefit is based upon wind power's energy and capacity services and the avoidance of environmental damages. The approach uses probabilistic modeling techniques to account for the stochastic interaction between wind power availability, electricity demand, and conventional generator dispatch. A method for including the spatial smoothing effect of geographically dispersed wind farms is also introduced. The model has been used to analyze potential offshore wind power development to the south of Long Island, NY. If natural gas combined cycle (NGCC) and integrated gasifier combined cycle (IGCC) are the alternative generation sources, wind power exhibits a negative social benefit due to its high capacity cost and the relatively low emissions of these advanced fossil-fuel technologies. Environmental benefits increase significantly if charges for CO2 emissions are included. Results also reveal a diminishing social benefit as wind power penetration increases. The dependence of wind power benefits on natural gas and coal prices is also discussed. In power systems with a high penetration of wind generated electricity, the intermittent availability of wind power may influence hourly spot prices. A price responsive electricity demand model is introduced that shows a small increase in wind power value when consumers react to hourly spot prices. The effectiveness of this mechanism depends heavily on estimates of the own- and cross-price elasticities of aggregate electricity demand. This work makes a valuable contribution by synthesizing information from research in power market economics, power system reliability, and environmental impact assessment, to develop a comprehensive methodology for analyzing wind power in the context of long-term energy planning.

Probabilistic brain tissue segmentation in neonatal magnetic resonance imaging.

PubMed

Anbeek, Petronella; Vincken, Koen L; Groenendaal, Floris; Koeman, Annemieke; van Osch, Matthias J P; van der Grond, Jeroen

2008-02-01

A fully automated method has been developed for segmentation of four different structures in the neonatal brain: white matter (WM), central gray matter (CEGM), cortical gray matter (COGM), and cerebrospinal fluid (CSF). The segmentation algorithm is based on information from T2-weighted (T2-w) and inversion recovery (IR) scans. The method uses a K nearest neighbor (KNN) classification technique with features derived from spatial information and voxel intensities. Probabilistic segmentations of each tissue type were generated. By applying thresholds on these probability maps, binary segmentations were obtained. These final segmentations were evaluated by comparison with a gold standard. The sensitivity, specificity, and Dice similarity index (SI) were calculated for quantitative validation of the results. High sensitivity and specificity with respect to the gold standard were reached: sensitivity >0.82 and specificity >0.9 for all tissue types. Tissue volumes were calculated from the binary and probabilistic segmentations. The probabilistic segmentation volumes of all tissue types accurately estimated the gold standard volumes. The KNN approach offers valuable ways for neonatal brain segmentation. The probabilistic outcomes provide a useful tool for accurate volume measurements. The described method is based on routine diagnostic magnetic resonance imaging (MRI) and is suitable for large population studies.

Predicting intensity ranks of peptide fragment ions.

PubMed

Frank, Ari M

2009-05-01

Accurate modeling of peptide fragmentation is necessary for the development of robust scoring functions for peptide-spectrum matches, which are the cornerstone of MS/MS-based identification algorithms. Unfortunately, peptide fragmentation is a complex process that can involve several competing chemical pathways, which makes it difficult to develop generative probabilistic models that describe it accurately. However, the vast amounts of MS/MS data being generated now make it possible to use data-driven machine learning methods to develop discriminative ranking-based models that predict the intensity ranks of a peptide's fragment ions. We use simple sequence-based features that get combined by a boosting algorithm into models that make peak rank predictions with high accuracy. In an accompanying manuscript, we demonstrate how these prediction models are used to significantly improve the performance of peptide identification algorithms. The models can also be useful in the design of optimal multiple reaction monitoring (MRM) transitions, in cases where there is insufficient experimental data to guide the peak selection process. The prediction algorithm can also be run independently through PepNovo+, which is available for download from http://bix.ucsd.edu/Software/PepNovo.html.

Predicting Intensity Ranks of Peptide Fragment Ions

PubMed Central

Frank, Ari M.

2009-01-01

Accurate modeling of peptide fragmentation is necessary for the development of robust scoring functions for peptide-spectrum matches, which are the cornerstone of MS/MS-based identification algorithms. Unfortunately, peptide fragmentation is a complex process that can involve several competing chemical pathways, which makes it difficult to develop generative probabilistic models that describe it accurately. However, the vast amounts of MS/MS data being generated now make it possible to use data-driven machine learning methods to develop discriminative ranking-based models that predict the intensity ranks of a peptide's fragment ions. We use simple sequence-based features that get combined by a boosting algorithm in to models that make peak rank predictions with high accuracy. In an accompanying manuscript, we demonstrate how these prediction models are used to significantly improve the performance of peptide identification algorithms. The models can also be useful in the design of optimal MRM transitions, in cases where there is insufficient experimental data to guide the peak selection process. The prediction algorithm can also be run independently through PepNovo+, which is available for download from http://bix.ucsd.edu/Software/PepNovo.html. PMID:19256476

Differential theory of learning for efficient neural network pattern recognition

NASA Astrophysics Data System (ADS)

Hampshire, John B., II; Vijaya Kumar, Bhagavatula

1993-08-01

We describe a new theory of differential learning by which a broad family of pattern classifiers (including many well-known neural network paradigms) can learn stochastic concepts efficiently. We describe the relationship between a classifier's ability to generalize well to unseen test examples and the efficiency of the strategy by which it learns. We list a series of proofs that differential learning is efficient in its information and computational resource requirements, whereas traditional probabilistic learning strategies are not. The proofs are illustrated by a simple example that lends itself to closed-form analysis. We conclude with an optical character recognition task for which three different types of differentially generated classifiers generalize significantly better than their probabilistically generated counterparts.

Using NCAR Yellowstone for PhotoVoltaic Power Forecasts with Artificial Neural Networks and an Analog Ensemble

NASA Astrophysics Data System (ADS)

Cervone, G.; Clemente-Harding, L.; Alessandrini, S.; Delle Monache, L.

2016-12-01

A methodology based on Artificial Neural Networks (ANN) and an Analog Ensemble (AnEn) is presented to generate 72-hour deterministic and probabilistic forecasts of power generated by photovoltaic (PV) power plants using input from a numerical weather prediction model and computed astronomical variables. ANN and AnEn are used individually and in combination to generate forecasts for three solar power plant located in Italy. The computational scalability of the proposed solution is tested using synthetic data simulating 4,450 PV power stations. The NCAR Yellowstone supercomputer is employed to test the parallel implementation of the proposed solution, ranging from 1 node (32 cores) to 4,450 nodes (141,140 cores). Results show that a combined AnEn + ANN solution yields best results, and that the proposed solution is well suited for massive scale computation.

On the use of faults and background seismicity in Seismic Probabilistic Tsunami Hazard Analysis (SPTHA)

NASA Astrophysics Data System (ADS)

Selva, Jacopo; Lorito, Stefano; Basili, Roberto; Tonini, Roberto; Tiberti, Mara Monica; Romano, Fabrizio; Perfetti, Paolo; Volpe, Manuela

2017-04-01

Most of the SPTHA studies and applications rely on several working assumptions: i) the - mostly offshore - tsunamigenic faults are sufficiently well known; ii) the subduction zone earthquakes dominate the hazard; iii) and their location and geometry is sufficiently well constrained. Hence, a probabilistic model is constructed as regards the magnitude-frequency distribution and sometimes the slip distribution of earthquakes occurring on assumed known faults. Then, tsunami scenarios are usually constructed for all earthquakes location, sizes, and slip distributions included in the probabilistic model, through deterministic numerical modelling of tsunami generation, propagation and impact on realistic bathymetries. Here, we adopt a different approach (Selva et al., GJI, 2016) that releases some of the above assumptions, considering that i) also non-subduction earthquakes may contribute significantly to SPTHA, depending on the local tectonic context; ii) that not all the offshore faults are known or sufficiently well constrained; iii) and that the faulting mechanism of future earthquakes cannot be considered strictly predictable. This approach uses as much as possible information from known faults which, depending on the amount of available information and on the local tectonic complexity, among other things, are either modelled as Predominant Seismicity (PS) or as Background Seismicity (BS). PS is used when it is possible to assume sufficiently known geometry and mechanism (e.g. for the main subduction zones). Conversely, within the BS approach information on faults is merged with that on past seismicity, dominant stress regime, and tectonic characterisation, to determine a probability density function for the faulting mechanism. To illustrate the methodology and its impact on the hazard estimates, we present an application in the NEAM region (Northeast Atlantic, Mediterranean and connected seas), initially designed during the ASTARTE project and now applied for the regional-scale SPTHA in the TSUMAPS-NEAM project funded by DG-ECHO.

Probabilistic inversion of expert assessments to inform projections about Antarctic ice sheet responses

PubMed Central

Wong, Tony E.; Keller, Klaus

2017-01-01

The response of the Antarctic ice sheet (AIS) to changing global temperatures is a key component of sea-level projections. Current projections of the AIS contribution to sea-level changes are deeply uncertain. This deep uncertainty stems, in part, from (i) the inability of current models to fully resolve key processes and scales, (ii) the relatively sparse available data, and (iii) divergent expert assessments. One promising approach to characterizing the deep uncertainty stemming from divergent expert assessments is to combine expert assessments, observations, and simple models by coupling probabilistic inversion and Bayesian inversion. Here, we present a proof-of-concept study that uses probabilistic inversion to fuse a simple AIS model and diverse expert assessments. We demonstrate the ability of probabilistic inversion to infer joint prior probability distributions of model parameters that are consistent with expert assessments. We then confront these inferred expert priors with instrumental and paleoclimatic observational data in a Bayesian inversion. These additional constraints yield tighter hindcasts and projections. We use this approach to quantify how the deep uncertainty surrounding expert assessments affects the joint probability distributions of model parameters and future projections. PMID:29287095

Efficient feature subset selection with probabilistic distance criteria. [pattern recognition

NASA Technical Reports Server (NTRS)

Chittineni, C. B.

1979-01-01

Recursive expressions are derived for efficiently computing the commonly used probabilistic distance measures as a change in the criteria both when a feature is added to and when a feature is deleted from the current feature subset. A combinatorial algorithm for generating all possible r feature combinations from a given set of s features in (s/r) steps with a change of a single feature at each step is presented. These expressions can also be used for both forward and backward sequential feature selection.

A Probabilistic Assessment of Failure for Air Force Building Systems

DTIC Science & Technology

2015-03-26

Rain Water Drainage System 1.000 0.225 0.085 0.424 0.800 0.968 0.998 4.449 D209001 Special Piping Systems 0.436 0.088 0.289 0.881 0.998 1.000 1.000...Rain Water Drainage 0.522 D2090 Other Plumbing Systems 0.303 D 30 H V A C D3010 Energy Supply 0.316 D3020 Heat Generating Systems 0.636 D3030...A PROBABILISTIC ASSESSMENT OF FAILURE FOR AIR FORCE BUILDING SYSTEMS THESIS Stephanie L

Uncertainty analysis of a groundwater flow model in East-central Florida.

PubMed

Sepúlveda, Nicasio; Doherty, John

2015-01-01

A groundwater flow model for east-central Florida has been developed to help water-resource managers assess the impact of increased groundwater withdrawals from the Floridan aquifer system on heads and spring flows originating from the Upper Floridan Aquifer. The model provides a probabilistic description of predictions of interest to water-resource managers, given the uncertainty associated with system heterogeneity, the large number of input parameters, and a nonunique groundwater flow solution. The uncertainty associated with these predictions can then be considered in decisions with which the model has been designed to assist. The "Null Space Monte Carlo" method is a stochastic probabilistic approach used to generate a suite of several hundred parameter field realizations, each maintaining the model in a calibrated state, and each considered to be hydrogeologically plausible. The results presented herein indicate that the model's capacity to predict changes in heads or spring flows that originate from increased groundwater withdrawals is considerably greater than its capacity to predict the absolute magnitudes of heads or spring flows. Furthermore, the capacity of the model to make predictions that are similar in location and in type to those in the calibration dataset exceeds its capacity to make predictions of different types at different locations. The quantification of these outcomes allows defensible use of the modeling process in support of future water-resources decisions. The model allows the decision-making process to recognize the uncertainties, and the spatial or temporal variability of uncertainties that are associated with predictions of future system behavior in a complex hydrogeological context. © 2014, National Ground Water Association.

Probabilistic Flood Mapping using Volunteered Geographical Information

NASA Astrophysics Data System (ADS)

Rivera, S. J.; Girons Lopez, M.; Seibert, J.; Minsker, B. S.

2016-12-01

Flood extent maps are widely used by decision makers and first responders to provide critical information that prevents economic impacts and the loss of human lives. These maps are usually obtained from sensory data and/or hydrologic models, which often have limited coverage in space and time. Recent developments in social media and communication technology have created a wealth of near-real-time, user-generated content during flood events in many urban areas, such as flooded locations, pictures of flooding extent and height, etc. These data could improve decision-making and response operations as events unfold. However, the integration of these data sources has been limited due to the need for methods that can extract and translate the data into useful information for decision-making. This study presents an approach that uses volunteer geographic information (VGI) and non-traditional data sources (i.e., Twitter, Flicker, YouTube, and 911 and 311 calls) to generate/update the flood extent maps in areas where no models and/or gauge data are operational. The approach combines Web-crawling and computer vision techniques to gather information about the location, extent, and water height of the flood from unstructured textual data, images, and videos. These estimates are then used to provide an updated flood extent map for areas surrounding the geo-coordinate of the VGI through the application of a Hydro Growing Region Algorithm (HGRA). HGRA combines hydrologic and image segmentation concepts to estimate a probabilistic flooding extent along the corresponding creeks. Results obtained for a case study in Austin, TX (i.e., 2015 Memorial Day flood) were comparable to those obtained by a calibrated hydrologic model and had good spatial correlation with flooding extents estimated by the Federal Emergency Management Agency (FEMA).

Application of the probabilistic approximate analysis method to a turbopump blade analysis. [for Space Shuttle Main Engine

NASA Technical Reports Server (NTRS)

Thacker, B. H.; Mcclung, R. C.; Millwater, H. R.

1990-01-01

An eigenvalue analysis of a typical space propulsion system turbopump blade is presented using an approximate probabilistic analysis methodology. The methodology was developed originally to investigate the feasibility of computing probabilistic structural response using closed-form approximate models. This paper extends the methodology to structures for which simple closed-form solutions do not exist. The finite element method will be used for this demonstration, but the concepts apply to any numerical method. The results agree with detailed analysis results and indicate the usefulness of using a probabilistic approximate analysis in determining efficient solution strategies.

Probabilistic Models for Solar Particle Events

NASA Technical Reports Server (NTRS)

Adams, James H., Jr.; Dietrich, W. F.; Xapsos, M. A.; Welton, A. M.

2009-01-01

Probabilistic Models of Solar Particle Events (SPEs) are used in space mission design studies to provide a description of the worst-case radiation environment that the mission must be designed to tolerate.The models determine the worst-case environment using a description of the mission and a user-specified confidence level that the provided environment will not be exceeded. This poster will focus on completing the existing suite of models by developing models for peak flux and event-integrated fluence elemental spectra for the Z>2 elements. It will also discuss methods to take into account uncertainties in the data base and the uncertainties resulting from the limited number of solar particle events in the database. These new probabilistic models are based on an extensive survey of SPE measurements of peak and event-integrated elemental differential energy spectra. Attempts are made to fit the measured spectra with eight different published models. The model giving the best fit to each spectrum is chosen and used to represent that spectrum for any energy in the energy range covered by the measurements. The set of all such spectral representations for each element is then used to determine the worst case spectrum as a function of confidence level. The spectral representation that best fits these worst case spectra is found and its dependence on confidence level is parameterized. This procedure creates probabilistic models for the peak and event-integrated spectra.

First USGS urban seismic hazard maps predict the effects of soils

USGS Publications Warehouse

Cramer, C.H.; Gomberg, J.S.; Schweig, E.S.; Waldron, B.A.; Tucker, K.

2006-01-01

Probabilistic and scenario urban seismic hazard maps have been produced for Memphis, Shelby County, Tennessee covering a six-quadrangle area of the city. The nine probabilistic maps are for peak ground acceleration and 0.2 s and 1.0 s spectral acceleration and for 10%, 5%, and 2% probability of being exceeded in 50 years. Six scenario maps for these three ground motions have also been generated for both an M7.7 and M6.2 on the southwest arm of the New Madrid seismic zone ending at Marked Tree, Arkansas. All maps include the effect of local geology. Relative to the national seismic hazard maps, the effect of the thick sediments beneath Memphis is to decrease 0.2 s probabilistic ground motions by 0-30% and increase 1.0 s probabilistic ground motions by ???100%. Probabilistic peak ground accelerations remain at levels similar to the national maps, although the ground motion gradient across Shelby County is reduced and ground motions are more uniform within the county. The M7.7 scenario maps show ground motions similar to the 5%-in-50-year probabilistic maps. As an effect of local geology, both M7.7 and M6.2 scenario maps show a more uniform seismic ground-motion hazard across Shelby County than scenario maps with constant site conditions (i.e., NEHRP B/C boundary).

Probabilistic evaluation of uncertainties and risks in aerospace components

NASA Technical Reports Server (NTRS)

Shah, A. R.; Shiao, M. C.; Nagpal, V. K.; Chamis, C. C.

1992-01-01

A methodology is presented for the computational simulation of primitive variable uncertainties, and attention is given to the simulation of specific aerospace components. Specific examples treated encompass a probabilistic material behavior model, as well as static, dynamic, and fatigue/damage analyses of a turbine blade in a mistuned bladed rotor in the SSME turbopumps. An account is given of the use of the NESSES probabilistic FEM analysis CFD code.

Using Bayesian Networks for Candidate Generation in Consistency-based Diagnosis

NASA Technical Reports Server (NTRS)

Narasimhan, Sriram; Mengshoel, Ole

2008-01-01

Consistency-based diagnosis relies heavily on the assumption that discrepancies between model predictions and sensor observations can be detected accurately. When sources of uncertainty like sensor noise and model abstraction exist robust schemes have to be designed to make a binary decision on whether predictions are consistent with observations. This risks the occurrence of false alarms and missed alarms when an erroneous decision is made. Moreover when multiple sensors (with differing sensing properties) are available the degree of match between predictions and observations can be used to guide the search for fault candidates. In this paper we propose a novel approach to handle this problem using Bayesian networks. In the consistency- based diagnosis formulation, automatically generated Bayesian networks are used to encode a probabilistic measure of fit between predictions and observations. A Bayesian network inference algorithm is used to compute most probable fault candidates.

Mathematical modeling of the malignancy of cancer using graph evolution.

PubMed

Gunduz-Demir, Cigdem

2007-10-01

We report a novel computational method based on graph evolution process to model the malignancy of brain cancer called glioma. In this work, we analyze the phases that a graph passes through during its evolution and demonstrate strong relation between the malignancy of cancer and the phase of its graph. From the photomicrographs of tissues, which are diagnosed as normal, low-grade cancerous and high-grade cancerous, we construct cell-graphs based on the locations of cells; we probabilistically generate an edge between every pair of cells depending on the Euclidean distance between them. For a cell-graph, we extract connectivity information including the properties of its connected components in order to analyze the phase of the cell-graph. Working with brain tissue samples surgically removed from 12 patients, we demonstrate that cell-graphs generated for different tissue types evolve differently and that they exhibit different phase properties, which distinguish a tissue type from another.

Probabilistic Design and Analysis Framework

NASA Technical Reports Server (NTRS)

Strack, William C.; Nagpal, Vinod K.

2010-01-01

PRODAF is a software package designed to aid analysts and designers in conducting probabilistic analysis of components and systems. PRODAF can integrate multiple analysis programs to ease the tedious process of conducting a complex analysis process that requires the use of multiple software packages. The work uses a commercial finite element analysis (FEA) program with modules from NESSUS to conduct a probabilistic analysis of a hypothetical turbine blade, disk, and shaft model. PRODAF applies the response surface method, at the component level, and extrapolates the component-level responses to the system level. Hypothetical components of a gas turbine engine are first deterministically modeled using FEA. Variations in selected geometrical dimensions and loading conditions are analyzed to determine the effects of the stress state within each component. Geometric variations include the cord length and height for the blade, inner radius, outer radius, and thickness, which are varied for the disk. Probabilistic analysis is carried out using developing software packages like System Uncertainty Analysis (SUA) and PRODAF. PRODAF was used with a commercial deterministic FEA program in conjunction with modules from the probabilistic analysis program, NESTEM, to perturb loads and geometries to provide a reliability and sensitivity analysis. PRODAF simplified the handling of data among the various programs involved, and will work with many commercial and opensource deterministic programs, probabilistic programs, or modules.

Sarma-based key-group method for rock slope reliability analyses

NASA Astrophysics Data System (ADS)

Yarahmadi Bafghi, A. R.; Verdel, T.

2005-08-01

The methods used in conducting static stability analyses have remained pertinent to this day for reasons of both simplicity and speed of execution. The most well-known of these methods for purposes of stability analysis of fractured rock masses is the key-block method (KBM).This paper proposes an extension to the KBM, called the key-group method (KGM), which combines not only individual key-blocks but also groups of collapsable blocks into an iterative and progressive analysis of the stability of discontinuous rock slopes. To take intra-group forces into account, the Sarma method has been implemented within the KGM in order to generate a Sarma-based KGM, abbreviated SKGM. We will discuss herein the hypothesis behind this new method, details regarding its implementation, and validation through comparison with results obtained from the distinct element method.Furthermore, as an alternative to deterministic methods, reliability analyses or probabilistic analyses have been proposed to take account of the uncertainty in analytical parameters and models. The FOSM and ASM probabilistic methods could be implemented within the KGM and SKGM framework in order to take account of the uncertainty due to physical and mechanical data (density, cohesion and angle of friction). We will then show how such reliability analyses can be introduced into SKGM to give rise to the probabilistic SKGM (PSKGM) and how it can be used for rock slope reliability analyses. Copyright

Source mechanisms of volcanic tsunamis.

PubMed

Paris, Raphaël

2015-10-28

Volcanic tsunamis are generated by a variety of mechanisms, including volcano-tectonic earthquakes, slope instabilities, pyroclastic flows, underwater explosions, shock waves and caldera collapse. In this review, we focus on the lessons that can be learnt from past events and address the influence of parameters such as volume flux of mass flows, explosion energy or duration of caldera collapse on tsunami generation. The diversity of waves in terms of amplitude, period, form, dispersion, etc. poses difficulties for integration and harmonization of sources to be used for numerical models and probabilistic tsunami hazard maps. In many cases, monitoring and warning of volcanic tsunamis remain challenging (further technical and scientific developments being necessary) and must be coupled with policies of population preparedness. © 2015 The Author(s).

Asteroid-Generated Tsunami and Impact Risk

NASA Astrophysics Data System (ADS)

Boslough, M.; Aftosmis, M.; Berger, M. J.; Ezzedine, S. M.; Gisler, G.; Jennings, B.; LeVeque, R. J.; Mathias, D.; McCoy, C.; Robertson, D.; Titov, V. V.; Wheeler, L.

2016-12-01

The justification for planetary defense comes from a cost/benefit analysis, which includes risk assessment. The contribution from ocean impacts and airbursts is difficult to quantify and represents a significant uncertainty in our assessment of the overall risk. Our group is currently working toward improved understanding of impact scenarios that can generate dangerous tsunami. The importance of asteroid-generated tsunami research has increased because a new Science Definition Team, at the behest of NASA's Planetary Defense Coordinating Office, is now updating the results of a 2003 study on which our current planetary defense policy is based Our group was formed to address this question on many fronts, including asteroid entry modeling, tsunami generation and propagation simulations, modeling of coastal run-ups, inundation, and consequences, infrastructure damage estimates, and physics-based probabilistic impact risk assessment. We also organized the Second International Workshop on Asteroid Threat Assessment, focused on asteroid-generated tsunami and associated risk (Aug. 23-24, 2016). We will summarize our progress and present the highlights of our workshop, emphasizing its relevance to earth and planetary science. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy under Contract DE-AC04-94AL85000.

Psychological Plausibility of the Theory of Probabilistic Mental Models and the Fast and Frugal Heuristics

ERIC Educational Resources Information Center

Dougherty, Michael R.; Franco-Watkins, Ana M.; Thomas, Rick

2008-01-01

The theory of probabilistic mental models (PMM; G. Gigerenzer, U. Hoffrage, & H. Kleinbolting, 1991) has had a major influence on the field of judgment and decision making, with the most recent important modifications to PMM theory being the identification of several fast and frugal heuristics (G. Gigerenzer & D. G. Goldstein, 1996). These…

Developing probabilistic models to predict amphibian site occupancy in a patchy landscape

Treesearch

R. A. Knapp; K.R. Matthews; H. K. Preisler; R. Jellison

2003-01-01

Abstract. Human-caused fragmentation of habitats is threatening an increasing number of animal and plant species, making an understanding of the factors influencing patch occupancy ever more important. The overall goal of the current study was to develop probabilistic models of patch occupancy for the mountain yellow-legged frog (Rana muscosa). This once-common species...

Probabilistic migration modelling focused on functional barrier efficiency and low migration concepts in support of risk assessment.

PubMed

Brandsch, Rainer

2017-10-01

Migration modelling provides reliable migration estimates from food-contact materials (FCM) to food or food simulants based on mass-transfer parameters like diffusion and partition coefficients related to individual materials. In most cases, mass-transfer parameters are not readily available from the literature and for this reason are estimated with a given uncertainty. Historically, uncertainty was accounted for by introducing upper limit concepts first, turning out to be of limited applicability due to highly overestimated migration results. Probabilistic migration modelling gives the possibility to consider uncertainty of the mass-transfer parameters as well as other model inputs. With respect to a functional barrier, the most important parameters among others are the diffusion properties of the functional barrier and its thickness. A software tool that accepts distribution as inputs and is capable of applying Monte Carlo methods, i.e., random sampling from the input distributions of the relevant parameters (i.e., diffusion coefficient and layer thickness), predicts migration results with related uncertainty and confidence intervals. The capabilities of probabilistic migration modelling are presented in the view of three case studies (1) sensitivity analysis, (2) functional barrier efficiency and (3) validation by experimental testing. Based on the predicted migration by probabilistic migration modelling and related exposure estimates, safety evaluation of new materials in the context of existing or new packaging concepts is possible. Identifying associated migration risk and potential safety concerns in the early stage of packaging development is possible. Furthermore, dedicated material selection exhibiting required functional barrier efficiency under application conditions becomes feasible. Validation of the migration risk assessment by probabilistic migration modelling through a minimum of dedicated experimental testing is strongly recommended.

Single, Complete, Probability Spaces Consistent With EPR-Bohm-Bell Experimental Data

NASA Astrophysics Data System (ADS)

Avis, David; Fischer, Paul; Hilbert, Astrid; Khrennikov, Andrei

2009-03-01

We show that paradoxical consequences of violations of Bell's inequality are induced by the use of an unsuitable probabilistic description for the EPR-Bohm-Bell experiment. The conventional description (due to Bell) is based on a combination of statistical data collected for different settings of polarization beam splitters (PBSs). In fact, such data consists of some conditional probabilities which only partially define a probability space. Ignoring this conditioning leads to apparent contradictions in the classical probabilistic model (due to Kolmogorov). We show how to make a completely consistent probabilistic model by taking into account the probabilities of selecting the settings of the PBSs. Our model matches both the experimental data and is consistent with classical probability theory.

Flood model for Brazil

NASA Astrophysics Data System (ADS)

Palán, Ladislav; Punčochář, Petr

2017-04-01

Looking on the impact of flooding from the World-wide perspective, in last 50 years flooding has caused over 460,000 fatalities and caused serious material damage. Combining economic loss from ten costliest flood events (from the same period) returns a loss (in the present value) exceeding 300bn USD. Locally, in Brazil, flood is the most damaging natural peril with alarming increase of events frequencies as 5 out of the 10 biggest flood losses ever recorded have occurred after 2009. The amount of economic and insured losses particularly caused by various flood types was the key driver of the local probabilistic flood model development. Considering the area of Brazil (being 5th biggest country in the World) and the scattered distribution of insured exposure, a domain covered by the model was limited to the entire state of Sao Paolo and 53 additional regions. The model quantifies losses on approx. 90 % of exposure (for regular property lines) of key insurers. Based on detailed exposure analysis, Impact Forecasting has developed this tool using long term local hydrological data series (Agencia Nacional de Aguas) from riverine gauge stations and digital elevation model (Instituto Brasileiro de Geografia e Estatística). To provide most accurate representation of local hydrological behaviour needed for the nature of probabilistic simulation, a hydrological data processing focused on frequency analyses of seasonal peak flows - done by fitting appropriate extreme value statistical distribution and stochastic event set generation consisting of synthetically derived flood events respecting realistic spatial and frequency patterns visible in entire period of hydrological observation. Data were tested for homogeneity, consistency and for any significant breakpoint occurrence in time series so the entire observation or only its subparts were used for further analysis. The realistic spatial patterns of stochastic events are reproduced through the innovative use of d-vine copula scheme to generate probabilistic flood event set. The derived design flows for selected rivers inside model domain were used as an input for 2-dimensional hydrodynamic inundation modelling techniques (using the tool TUFLOW by BMT WBM) on mesh size 30 x 30 metres. Outputs from inundation modelling and stochastic event set were implemented in the Aon Benfield's platform ELEMENTS developed and managed internally by Impact Forecasting; Aon Benfield internal catastrophe model development center. The model was designed to evaluate potential financial impact caused by fluvial flooding on portfolios of insurance and/or reinsurance companies. The structure of presented model follows typical scheme of financial loss catastrophe model and combines hazard with exposure and vulnerability to produce potential financial loss expressed in the form of loss exceedance probability curve and many other insured perspectives, such as average annual loss, event or quantile loss tables and etc. Model can take financial inputs as well as provide split of results for exact specified location or related higher administrative units: municipalities and 5-digit postal codes.

Infering and Calibrating Triadic Closure in a Dynamic Network

NASA Astrophysics Data System (ADS)

Mantzaris, Alexander V.; Higham, Desmond J.

In the social sciences, the hypothesis of triadic closure contends that new links in a social contact network arise preferentially between those who currently share neighbours. Here, in a proof-of-principle study, we show how to calibrate a recently proposed evolving network model to time-dependent connectivity data. The probabilistic edge birth rate in the model contains a triadic closure term, so we are also able to assess statistically the evidence for this effect. The approach is shown to work on data generated synthetically from the model. We then apply this methodology to some real, large-scale data that records the build up of connections in a business-related social networking site, and find evidence for triadic closure.

Probabilistic learning of nonlinear dynamical systems using sequential Monte Carlo

NASA Astrophysics Data System (ADS)

Schön, Thomas B.; Svensson, Andreas; Murray, Lawrence; Lindsten, Fredrik

2018-05-01

Probabilistic modeling provides the capability to represent and manipulate uncertainty in data, models, predictions and decisions. We are concerned with the problem of learning probabilistic models of dynamical systems from measured data. Specifically, we consider learning of probabilistic nonlinear state-space models. There is no closed-form solution available for this problem, implying that we are forced to use approximations. In this tutorial we will provide a self-contained introduction to one of the state-of-the-art methods-the particle Metropolis-Hastings algorithm-which has proven to offer a practical approximation. This is a Monte Carlo based method, where the particle filter is used to guide a Markov chain Monte Carlo method through the parameter space. One of the key merits of the particle Metropolis-Hastings algorithm is that it is guaranteed to converge to the "true solution" under mild assumptions, despite being based on a particle filter with only a finite number of particles. We will also provide a motivating numerical example illustrating the method using a modeling language tailored for sequential Monte Carlo methods. The intention of modeling languages of this kind is to open up the power of sophisticated Monte Carlo methods-including particle Metropolis-Hastings-to a large group of users without requiring them to know all the underlying mathematical details.

Ensemble Simulations with Coupled Atmospheric Dynamic and Dispersion Models: Illustrating Uncertainties in Dosage Simulations.

NASA Astrophysics Data System (ADS)

Warner, Thomas T.; Sheu, Rong-Shyang; Bowers, James F.; Sykes, R. Ian; Dodd, Gregory C.; Henn, Douglas S.

2002-05-01

Ensemble simulations made using a coupled atmospheric dynamic model and a probabilistic Lagrangian puff dispersion model were employed in a forensic analysis of the transport and dispersion of a toxic gas that may have been released near Al Muthanna, Iraq, during the Gulf War. The ensemble study had two objectives, the first of which was to determine the sensitivity of the calculated dosage fields to the choices that must be made about the configuration of the atmospheric dynamic model. In this test, various choices were used for model physics representations and for the large-scale analyses that were used to construct the model initial and boundary conditions. The second study objective was to examine the dispersion model's ability to use ensemble inputs to predict dosage probability distributions. Here, the dispersion model was used with the ensemble mean fields from the individual atmospheric dynamic model runs, including the variability in the individual wind fields, to generate dosage probabilities. These are compared with the explicit dosage probabilities derived from the individual runs of the coupled modeling system. The results demonstrate that the specific choices made about the dynamic-model configuration and the large-scale analyses can have a large impact on the simulated dosages. For example, the area near the source that is exposed to a selected dosage threshold varies by up to a factor of 4 among members of the ensemble. The agreement between the explicit and ensemble dosage probabilities is relatively good for both low and high dosage levels. Although only one ensemble was considered in this study, the encouraging results suggest that a probabilistic dispersion model may be of value in quantifying the effects of uncertainties in a dynamic-model ensemble on dispersion model predictions of atmospheric transport and dispersion.

Development and Validation of a Statistical Shape Modeling-Based Finite Element Model of the Cervical Spine Under Low-Level Multiple Direction Loading Conditions

PubMed Central

Bredbenner, Todd L.; Eliason, Travis D.; Francis, W. Loren; McFarland, John M.; Merkle, Andrew C.; Nicolella, Daniel P.

2014-01-01

Cervical spinal injuries are a significant concern in all trauma injuries. Recent military conflicts have demonstrated the substantial risk of spinal injury for the modern warfighter. Finite element models used to investigate injury mechanisms often fail to examine the effects of variation in geometry or material properties on mechanical behavior. The goals of this study were to model geometric variation for a set of cervical spines, to extend this model to a parametric finite element model, and, as a first step, to validate the parametric model against experimental data for low-loading conditions. Individual finite element models were created using cervical spine (C3–T1) computed tomography data for five male cadavers. Statistical shape modeling (SSM) was used to generate a parametric finite element model incorporating variability of spine geometry, and soft-tissue material property variation was also included. The probabilistic loading response of the parametric model was determined under flexion-extension, axial rotation, and lateral bending and validated by comparison to experimental data. Based on qualitative and quantitative comparison of the experimental loading response and model simulations, we suggest that the model performs adequately under relatively low-level loading conditions in multiple loading directions. In conclusion, SSM methods coupled with finite element analyses within a probabilistic framework, along with the ability to statistically validate the overall model performance, provide innovative and important steps toward describing the differences in vertebral morphology, spinal curvature, and variation in material properties. We suggest that these methods, with additional investigation and validation under injurious loading conditions, will lead to understanding and mitigating the risks of injury in the spine and other musculoskeletal structures. PMID:25506051

Uncertainty analysis of a groundwater flow model in east-central Florida

USGS Publications Warehouse

Sepúlveda, Nicasio; Doherty, John E.

2014-01-01

A groundwater flow model for east-central Florida has been developed to help water-resource managers assess the impact of increased groundwater withdrawals from the Floridan aquifer system on heads and spring flows originating from the Upper Floridan aquifer. The model provides a probabilistic description of predictions of interest to water-resource managers, given the uncertainty associated with system heterogeneity, the large number of input parameters, and a nonunique groundwater flow solution. The uncertainty associated with these predictions can then be considered in decisions with which the model has been designed to assist. The “Null Space Monte Carlo” method is a stochastic probabilistic approach used to generate a suite of several hundred parameter field realizations, each maintaining the model in a calibrated state, and each considered to be hydrogeologically plausible. The results presented herein indicate that the model’s capacity to predict changes in heads or spring flows that originate from increased groundwater withdrawals is considerably greater than its capacity to predict the absolute magnitudes of heads or spring flows. Furthermore, the capacity of the model to make predictions that are similar in location and in type to those in the calibration dataset exceeds its capacity to make predictions of different types at different locations. The quantification of these outcomes allows defensible use of the modeling process in support of future water-resources decisions. The model allows the decision-making process to recognize the uncertainties, and the spatial/temporal variability of uncertainties that are associated with predictions of future system behavior in a complex hydrogeological context.

Statistical learning and probabilistic prediction in music cognition: mechanisms of stylistic enculturation.

PubMed

Pearce, Marcus T

2018-05-11

Music perception depends on internal psychological models derived through exposure to a musical culture. It is hypothesized that this musical enculturation depends on two cognitive processes: (1) statistical learning, in which listeners acquire internal cognitive models of statistical regularities present in the music to which they are exposed; and (2) probabilistic prediction based on these learned models that enables listeners to organize and process their mental representations of music. To corroborate these hypotheses, I review research that uses a computational model of probabilistic prediction based on statistical learning (the information dynamics of music (IDyOM) model) to simulate data from empirical studies of human listeners. The results show that a broad range of psychological processes involved in music perception-expectation, emotion, memory, similarity, segmentation, and meter-can be understood in terms of a single, underlying process of probabilistic prediction using learned statistical models. Furthermore, IDyOM simulations of listeners from different musical cultures demonstrate that statistical learning can plausibly predict causal effects of differential cultural exposure to musical styles, providing a quantitative model of cultural distance. Understanding the neural basis of musical enculturation will benefit from close coordination between empirical neuroimaging and computational modeling of underlying mechanisms, as outlined here. © 2018 The Authors. Annals of the New York Academy of Sciences published by Wiley Periodicals, Inc. on behalf of New York Academy of Sciences.

A note on probabilistic models over strings: the linear algebra approach.

PubMed

Bouchard-Côté, Alexandre

2013-12-01

Probabilistic models over strings have played a key role in developing methods that take into consideration indels as phylogenetically informative events. There is an extensive literature on using automata and transducers on phylogenies to do inference on these probabilistic models, in which an important theoretical question is the complexity of computing the normalization of a class of string-valued graphical models. This question has been investigated using tools from combinatorics, dynamic programming, and graph theory, and has practical applications in Bayesian phylogenetics. In this work, we revisit this theoretical question from a different point of view, based on linear algebra. The main contribution is a set of results based on this linear algebra view that facilitate the analysis and design of inference algorithms on string-valued graphical models. As an illustration, we use this method to give a new elementary proof of a known result on the complexity of inference on the "TKF91" model, a well-known probabilistic model over strings. Compared to previous work, our proving method is easier to extend to other models, since it relies on a novel weak condition, triangular transducers, which is easy to establish in practice. The linear algebra view provides a concise way of describing transducer algorithms and their compositions, opens the possibility of transferring fast linear algebra libraries (for example, based on GPUs), as well as low rank matrix approximation methods, to string-valued inference problems.

Performance assessment of deterministic and probabilistic weather predictions for the short-term optimization of a tropical hydropower reservoir

NASA Astrophysics Data System (ADS)

Mainardi Fan, Fernando; Schwanenberg, Dirk; Alvarado, Rodolfo; Assis dos Reis, Alberto; Naumann, Steffi; Collischonn, Walter

2016-04-01

Hydropower is the most important electricity source in Brazil. During recent years, it accounted for 60% to 70% of the total electric power supply. Marginal costs of hydropower are lower than for thermal power plants, therefore, there is a strong economic motivation to maximize its share. On the other hand, hydropower depends on the availability of water, which has a natural variability. Its extremes lead to the risks of power production deficits during droughts and safety issues in the reservoir and downstream river reaches during flood events. One building block of the proper management of hydropower assets is the short-term forecast of reservoir inflows as input for an online, event-based optimization of its release strategy. While deterministic forecasts and optimization schemes are the established techniques for the short-term reservoir management, the use of probabilistic ensemble forecasts and stochastic optimization techniques receives growing attention and a number of researches have shown its benefit. The present work shows one of the first hindcasting and closed-loop control experiments for a multi-purpose hydropower reservoir in a tropical region in Brazil. The case study is the hydropower project (HPP) Três Marias, located in southeast Brazil. The HPP reservoir is operated with two main objectives: (i) hydroelectricity generation and (ii) flood control at Pirapora City located 120 km downstream of the dam. In the experiments, precipitation forecasts based on observed data, deterministic and probabilistic forecasts with 50 ensemble members of the ECMWF are used as forcing of the MGB-IPH hydrological model to generate streamflow forecasts over a period of 2 years. The online optimization depends on a deterministic and multi-stage stochastic version of a model predictive control scheme. Results for the perfect forecasts show the potential benefit of the online optimization and indicate a desired forecast lead time of 30 days. In comparison, the use of actual forecasts with shorter lead times of up to 15 days shows the practical benefit of actual operational data. It appears that the use of stochastic optimization combined with ensemble forecasts leads to a significant higher level of flood protection without compromising the HPP's energy production.

Using Analog Ensemble to generate spatially downscaled probabilistic wind power forecasts

NASA Astrophysics Data System (ADS)

Delle Monache, L.; Shahriari, M.; Cervone, G.

2017-12-01

We use the Analog Ensemble (AnEn) method to generate probabilistic 80-m wind power forecasts. We use data from the NCEP GFS ( 28 km resolution) and NCEP NAM (12 km resolution). We use forecasts data from NAM and GFS, and analysis data from NAM which enables us to: 1) use a lower-resolution model to create higher-resolution forecasts, and 2) use a higher-resolution model to create higher-resolution forecasts. The former essentially increases computing speed and the latter increases forecast accuracy. An aggregated model of the former can be compared against the latter to measure the accuracy of the AnEn spatial downscaling. The AnEn works by taking a deterministic future forecast and comparing it with past forecasts. The model searches for the best matching estimates within the past forecasts and selects the predictand value corresponding to these past forecasts as the ensemble prediction for the future forecast. Our study is based on predicting wind speed and air density at more than 13,000 grid points in the continental US. We run the AnEn model twice: 1) estimating 80-m wind speed by using predictor variables such as temperature, pressure, geopotential height, U-component and V-component of wind, 2) estimating air density by using predictors such as temperature, pressure, and relative humidity. We use the air density values to correct the standard wind power curves for different values of air density. The standard deviation of the ensemble members (i.e. ensemble spread) will be used as the degree of difficulty to predict wind power at different locations. The value of the correlation coefficient between the ensemble spread and the forecast error determines the appropriateness of this measure. This measure is prominent for wind farm developers as building wind farms in regions with higher predictability will reduce the real-time risks of operating in the electricity markets.

Seismic Risk Assessment for the Kyrgyz Republic

NASA Astrophysics Data System (ADS)

Pittore, Massimiliano; Sousa, Luis; Grant, Damian; Fleming, Kevin; Parolai, Stefano; Fourniadis, Yannis; Free, Matthew; Moldobekov, Bolot; Takeuchi, Ko

2017-04-01

The Kyrgyz Republic is one of the most socially and economically dynamic countries in Central Asia, and one of the most endangered by earthquake hazard in the region. In order to support the government of the Kyrgyz Republic in the development of a country-level Disaster Risk Reduction strategy, a comprehensive seismic risk study has been developed with the support of the World Bank. As part of this project, state-of-the-art hazard, exposure and vulnerability models have been developed and combined into the assessment of direct physical and economic risk on residential, educational and transportation infrastructure. The seismic hazard has been modelled with three different approaches, in order to provide a comprehensive overview of the possible consequences. A probabilistic seismic hazard assessment (PSHA) approach has been used to quantitatively evaluate the distribution of expected ground shaking intensity, as constrained by the compiled earthquake catalogue and associated seismic source model. A set of specific seismic scenarios based on events generated from known fault systems have been also considered, in order to provide insight on the expected consequences in case of strong events in proximity of densely inhabited areas. Furthermore, long-span catalogues of events have been generated stochastically and employed in the probabilistic analysis of expected losses over the territory of the Kyrgyz Republic. Damage and risk estimates have been computed by using an exposure model recently developed for the country, combined with the assignment of suitable fragility/vulnerability models. The risk estimation has been carried out with spatial aggregation at the district (rayon) level. The obtained results confirm the high level of seismic risk throughout the country, also pinpointing the location of several risk hotspots, particularly in the southern districts, in correspondence with the Ferghana valley. The outcome of this project will further support the local decision makers in implementing specific prevention and mitigation measures that are consistent with a broad risk reduction strategy.

Unsupervised, low latency anomaly detection of algorithmically generated domain names by generative probabilistic modeling.

PubMed

Raghuram, Jayaram; Miller, David J; Kesidis, George

2014-07-01

We propose a method for detecting anomalous domain names, with focus on algorithmically generated domain names which are frequently associated with malicious activities such as fast flux service networks, particularly for bot networks (or botnets), malware, and phishing. Our method is based on learning a (null hypothesis) probability model based on a large set of domain names that have been white listed by some reliable authority. Since these names are mostly assigned by humans, they are pronounceable, and tend to have a distribution of characters, words, word lengths, and number of words that are typical of some language (mostly English), and often consist of words drawn from a known lexicon. On the other hand, in the present day scenario, algorithmically generated domain names typically have distributions that are quite different from that of human-created domain names. We propose a fully generative model for the probability distribution of benign (white listed) domain names which can be used in an anomaly detection setting for identifying putative algorithmically generated domain names. Unlike other methods, our approach can make detections without considering any additional (latency producing) information sources, often used to detect fast flux activity. Experiments on a publicly available, large data set of domain names associated with fast flux service networks show encouraging results, relative to several baseline methods, with higher detection rates and low false positive rates.

Unsupervised, low latency anomaly detection of algorithmically generated domain names by generative probabilistic modeling

PubMed Central

Raghuram, Jayaram; Miller, David J.; Kesidis, George

2014-01-01

We propose a method for detecting anomalous domain names, with focus on algorithmically generated domain names which are frequently associated with malicious activities such as fast flux service networks, particularly for bot networks (or botnets), malware, and phishing. Our method is based on learning a (null hypothesis) probability model based on a large set of domain names that have been white listed by some reliable authority. Since these names are mostly assigned by humans, they are pronounceable, and tend to have a distribution of characters, words, word lengths, and number of words that are typical of some language (mostly English), and often consist of words drawn from a known lexicon. On the other hand, in the present day scenario, algorithmically generated domain names typically have distributions that are quite different from that of human-created domain names. We propose a fully generative model for the probability distribution of benign (white listed) domain names which can be used in an anomaly detection setting for identifying putative algorithmically generated domain names. Unlike other methods, our approach can make detections without considering any additional (latency producing) information sources, often used to detect fast flux activity. Experiments on a publicly available, large data set of domain names associated with fast flux service networks show encouraging results, relative to several baseline methods, with higher detection rates and low false positive rates. PMID:25685511

An analytical probabilistic model of the quality efficiency of a sewer tank

NASA Astrophysics Data System (ADS)

Balistrocchi, Matteo; Grossi, Giovanna; Bacchi, Baldassare

2009-12-01

The assessment of the efficiency of a storm water storage facility devoted to the sewer overflow control in urban areas strictly depends on the ability to model the main features of the rainfall-runoff routing process and the related wet weather pollution delivery. In this paper the possibility of applying the analytical probabilistic approach for developing a tank design method, whose potentials are similar to the continuous simulations, is proved. In the model derivation the quality issues of such devices were implemented. The formulation is based on a Weibull probabilistic model of the main characteristics of the rainfall process and on a power law describing the relationship between the dimensionless storm water cumulative runoff volume and the dimensionless cumulative pollutograph. Following this approach, efficiency indexes were established. The proposed model was verified by comparing its results to those obtained by continuous simulations; satisfactory agreement is shown for the proposed efficiency indexes.

Towards a high-speed quantum random number generator

NASA Astrophysics Data System (ADS)

Stucki, Damien; Burri, Samuel; Charbon, Edoardo; Chunnilall, Christopher; Meneghetti, Alessio; Regazzoni, Francesco

2013-10-01

Randomness is of fundamental importance in various fields, such as cryptography, numerical simulations, or the gaming industry. Quantum physics, which is fundamentally probabilistic, is the best option for a physical random number generator. In this article, we will present the work carried out in various projects in the context of the development of a commercial and certified high speed random number generator.

PROBABILISTIC MODELING FOR ADVANCED HUMAN EXPOSURE ASSESSMENT

EPA Science Inventory

Human exposures to environmental pollutants widely vary depending on the emission patterns that result in microenvironmental pollutant concentrations, as well as behavioral factors that determine the extent of an individual's contact with these pollutants. Probabilistic human exp...

Exploiting the functional and taxonomic structure of genomic data by probabilistic topic modeling.

PubMed

Chen, Xin; Hu, Xiaohua; Lim, Tze Y; Shen, Xiajiong; Park, E K; Rosen, Gail L

2012-01-01

In this paper, we present a method that enable both homology-based approach and composition-based approach to further study the functional core (i.e., microbial core and gene core, correspondingly). In the proposed method, the identification of major functionality groups is achieved by generative topic modeling, which is able to extract useful information from unlabeled data. We first show that generative topic model can be used to model the taxon abundance information obtained by homology-based approach and study the microbial core. The model considers each sample as a “document,” which has a mixture of functional groups, while each functional group (also known as a “latent topic”) is a weight mixture of species. Therefore, estimating the generative topic model for taxon abundance data will uncover the distribution over latent functions (latent topic) in each sample. Second, we show that, generative topic model can also be used to study the genome-level composition of “N-mer” features (DNA subreads obtained by composition-based approaches). The model consider each genome as a mixture of latten genetic patterns (latent topics), while each functional pattern is a weighted mixture of the “N-mer” features, thus the existence of core genomes can be indicated by a set of common N-mer features. After studying the mutual information between latent topics and gene regions, we provide an explanation of the functional roles of uncovered latten genetic patterns. The experimental results demonstrate the effectiveness of proposed method.

Probabilistic boundary element method

NASA Technical Reports Server (NTRS)

Cruse, T. A.; Raveendra, S. T.

1989-01-01

The purpose of the Probabilistic Structural Analysis Method (PSAM) project is to develop structural analysis capabilities for the design analysis of advanced space propulsion system hardware. The boundary element method (BEM) is used as the basis of the Probabilistic Advanced Analysis Methods (PADAM) which is discussed. The probabilistic BEM code (PBEM) is used to obtain the structural response and sensitivity results to a set of random variables. As such, PBEM performs analogous to other structural analysis codes such as finite elements in the PSAM system. For linear problems, unlike the finite element method (FEM), the BEM governing equations are written at the boundary of the body only, thus, the method eliminates the need to model the volume of the body. However, for general body force problems, a direct condensation of the governing equations to the boundary of the body is not possible and therefore volume modeling is generally required.

A probabilistic approach to radiative energy loss calculations for optically thick atmospheres - Hydrogen lines and continua

NASA Technical Reports Server (NTRS)

Canfield, R. C.; Ricchiazzi, P. J.

1980-01-01

An approximate probabilistic radiative transfer equation and the statistical equilibrium equations are simultaneously solved for a model hydrogen atom consisting of three bound levels and ionization continuum. The transfer equation for L-alpha, L-beta, H-alpha, and the Lyman continuum is explicitly solved assuming complete redistribution. The accuracy of this approach is tested by comparing source functions and radiative loss rates to values obtained with a method that solves the exact transfer equation. Two recent model solar-flare chromospheres are used for this test. It is shown that for the test atmospheres the probabilistic method gives values of the radiative loss rate that are characteristically good to a factor of 2. The advantage of this probabilistic approach is that it retains a description of the dominant physical processes of radiative transfer in the complete redistribution case, yet it achieves a major reduction in computational requirements.

Quantification and Segmentation of Brain Tissues from MR Images: A Probabilistic Neural Network Approach

PubMed Central

Wang, Yue; Adalý, Tülay; Kung, Sun-Yuan; Szabo, Zsolt

2007-01-01

This paper presents a probabilistic neural network based technique for unsupervised quantification and segmentation of brain tissues from magnetic resonance images. It is shown that this problem can be solved by distribution learning and relaxation labeling, resulting in an efficient method that may be particularly useful in quantifying and segmenting abnormal brain tissues where the number of tissue types is unknown and the distributions of tissue types heavily overlap. The new technique uses suitable statistical models for both the pixel and context images and formulates the problem in terms of model-histogram fitting and global consistency labeling. The quantification is achieved by probabilistic self-organizing mixtures and the segmentation by a probabilistic constraint relaxation network. The experimental results show the efficient and robust performance of the new algorithm and that it outperforms the conventional classification based approaches. PMID:18172510

Time Alignment as a Necessary Step in the Analysis of Sleep Probabilistic Curves

NASA Astrophysics Data System (ADS)

Rošt'áková, Zuzana; Rosipal, Roman

2018-02-01

Sleep can be characterised as a dynamic process that has a finite set of sleep stages during the night. The standard Rechtschaffen and Kales sleep model produces discrete representation of sleep and does not take into account its dynamic structure. In contrast, the continuous sleep representation provided by the probabilistic sleep model accounts for the dynamics of the sleep process. However, analysis of the sleep probabilistic curves is problematic when time misalignment is present. In this study, we highlight the necessity of curve synchronisation before further analysis. Original and in time aligned sleep probabilistic curves were transformed into a finite dimensional vector space, and their ability to predict subjects' age or daily measures is evaluated. We conclude that curve alignment significantly improves the prediction of the daily measures, especially in the case of the S2-related sleep states or slow wave sleep.

Interpreting intraplate tectonics for seismic hazard: a UK historical perspective

NASA Astrophysics Data System (ADS)

Musson, R. M. W.

2012-04-01

It is notoriously difficult to construct seismic source models for probabilistic seismic hazard assessment in intraplate areas on the basis of geological information, and many practitioners have given up the task in favour of purely seismicity-based models. This risks losing potentially valuable information in regions where the earthquake catalogue is short compared to the seismic cycle. It is interesting to survey how attitudes to this issue have evolved over the past 30 years. This paper takes the UK as an example, and traces the evolution of seismic source models through generations of hazard studies. It is found that in the UK, while the earliest studies did not consider regional tectonics in any way, there has been a gradual evolution towards more tectonically based models. Experience in other countries, of course, may differ.

A fortran program for Monte Carlo simulation of oil-field discovery sequences

USGS Publications Warehouse

Bohling, Geoffrey C.; Davis, J.C.

1993-01-01

We have developed a program for performing Monte Carlo simulation of oil-field discovery histories. A synthetic parent population of fields is generated as a finite sample from a distribution of specified form. The discovery sequence then is simulated by sampling without replacement from this parent population in accordance with a probabilistic discovery process model. The program computes a chi-squared deviation between synthetic and actual discovery sequences as a function of the parameters of the discovery process model, the number of fields in the parent population, and the distributional parameters of the parent population. The program employs the three-parameter log gamma model for the distribution of field sizes and employs a two-parameter discovery process model, allowing the simulation of a wide range of scenarios. ?? 1993.

An open-access CMIP5 pattern library for temperature and precipitation: Description and methodology

DOE Office of Scientific and Technical Information (OSTI.GOV)

Lynch, Cary D.; Hartin, Corinne A.; Bond-Lamberty, Benjamin

Pattern scaling is used to efficiently emulate general circulation models and explore uncertainty in climate projections under multiple forcing scenarios. Pattern scaling methods assume that local climate changes scale with a global mean temperature increase, allowing for spatial patterns to be generated for multiple models for any future emission scenario. For uncertainty quantification and probabilistic statistical analysis, a library of patterns with descriptive statistics for each file would be beneficial, but such a library does not presently exist. Of the possible techniques used to generate patterns, the two most prominent are the delta and least squared regression methods. We exploremore » the differences and statistical significance between patterns generated by each method and assess performance of the generated patterns across methods and scenarios. Differences in patterns across seasons between methods and epochs were largest in high latitudes (60-90°N/S). Bias and mean errors between modeled and pattern predicted output from the linear regression method were smaller than patterns generated by the delta method. Across scenarios, differences in the linear regression method patterns were more statistically significant, especially at high latitudes. We found that pattern generation methodologies were able to approximate the forced signal of change to within ≤ 0.5°C, but choice of pattern generation methodology for pattern scaling purposes should be informed by user goals and criteria. As a result, this paper describes our library of least squared regression patterns from all CMIP5 models for temperature and precipitation on an annual and sub-annual basis, along with the code used to generate these patterns.« less

An open-access CMIP5 pattern library for temperature and precipitation: Description and methodology

DOE PAGES

Lynch, Cary D.; Hartin, Corinne A.; Bond-Lamberty, Benjamin; ...

2017-05-15

Pattern scaling is used to efficiently emulate general circulation models and explore uncertainty in climate projections under multiple forcing scenarios. Pattern scaling methods assume that local climate changes scale with a global mean temperature increase, allowing for spatial patterns to be generated for multiple models for any future emission scenario. For uncertainty quantification and probabilistic statistical analysis, a library of patterns with descriptive statistics for each file would be beneficial, but such a library does not presently exist. Of the possible techniques used to generate patterns, the two most prominent are the delta and least squared regression methods. We exploremore » the differences and statistical significance between patterns generated by each method and assess performance of the generated patterns across methods and scenarios. Differences in patterns across seasons between methods and epochs were largest in high latitudes (60-90°N/S). Bias and mean errors between modeled and pattern predicted output from the linear regression method were smaller than patterns generated by the delta method. Across scenarios, differences in the linear regression method patterns were more statistically significant, especially at high latitudes. We found that pattern generation methodologies were able to approximate the forced signal of change to within ≤ 0.5°C, but choice of pattern generation methodology for pattern scaling purposes should be informed by user goals and criteria. As a result, this paper describes our library of least squared regression patterns from all CMIP5 models for temperature and precipitation on an annual and sub-annual basis, along with the code used to generate these patterns.« less

Leveraging Past and Current Measurements to Probabilistically Nowcast Low Visibility Procedures at an Airport

NASA Astrophysics Data System (ADS)

Mayr, G. J.; Kneringer, P.; Dietz, S. J.; Zeileis, A.

2016-12-01

Low visibility or low cloud ceiling reduce the capacity of airports by requiring special low visibility procedures (LVP) for incoming/departing aircraft. Probabilistic forecasts when such procedures will become necessary help to mitigate delays and economic losses.We compare the performance of probabilistic nowcasts with two statistical methods: ordered logistic regression, and trees and random forests. These models harness historic and current meteorological measurements in the vicinity of the airport and LVP states, and incorporate diurnal and seasonal climatological information via generalized additive models (GAM). The methods are applied at Vienna International Airport (Austria). The performance is benchmarked against climatology, persistence and human forecasters.

A DATA-DRIVEN MODEL FOR SPECTRA: FINDING DOUBLE REDSHIFTS IN THE SLOAN DIGITAL SKY SURVEY

DOE Office of Scientific and Technical Information (OSTI.GOV)

Tsalmantza, P.; Hogg, David W., E-mail: vivitsal@mpia.de

2012-07-10

We present a data-driven method-heteroscedastic matrix factorization, a kind of probabilistic factor analysis-for modeling or performing dimensionality reduction on observed spectra or other high-dimensional data with known but non-uniform observational uncertainties. The method uses an iterative inverse-variance-weighted least-squares minimization procedure to generate a best set of basis functions. The method is similar to principal components analysis (PCA), but with the substantial advantage that it uses measurement uncertainties in a responsible way and accounts naturally for poorly measured and missing data; it models the variance in the noise-deconvolved data space. A regularization can be applied, in the form of a smoothnessmore » prior (inspired by Gaussian processes) or a non-negative constraint, without making the method prohibitively slow. Because the method optimizes a justified scalar (related to the likelihood), the basis provides a better fit to the data in a probabilistic sense than any PCA basis. We test the method on Sloan Digital Sky Survey (SDSS) spectra, concentrating on spectra known to contain two redshift components: these are spectra of gravitational lens candidates and massive black hole binaries. We apply a hypothesis test to compare one-redshift and two-redshift models for these spectra, utilizing the data-driven model trained on a random subset of all SDSS spectra. This test confirms 129 of the 131 lens candidates in our sample and all of the known binary candidates, and turns up very few false positives.« less

Simultaneous calibration of ensemble river flow predictions over an entire range of lead times

NASA Astrophysics Data System (ADS)

Hemri, S.; Fundel, F.; Zappa, M.

2013-10-01

Probabilistic estimates of future water levels and river discharge are usually simulated with hydrologic models using ensemble weather forecasts as main inputs. As hydrologic models are imperfect and the meteorological ensembles tend to be biased and underdispersed, the ensemble forecasts for river runoff typically are biased and underdispersed, too. Thus, in order to achieve both reliable and sharp predictions statistical postprocessing is required. In this work Bayesian model averaging (BMA) is applied to statistically postprocess ensemble runoff raw forecasts for a catchment in Switzerland, at lead times ranging from 1 to 240 h. The raw forecasts have been obtained using deterministic and ensemble forcing meteorological models with different forecast lead time ranges. First, BMA is applied based on mixtures of univariate normal distributions, subject to the assumption of independence between distinct lead times. Then, the independence assumption is relaxed in order to estimate multivariate runoff forecasts over the entire range of lead times simultaneously, based on a BMA version that uses multivariate normal distributions. Since river runoff is a highly skewed variable, Box-Cox transformations are applied in order to achieve approximate normality. Both univariate and multivariate BMA approaches are able to generate well calibrated probabilistic forecasts that are considerably sharper than climatological forecasts. Additionally, multivariate BMA provides a promising approach for incorporating temporal dependencies into the postprocessed forecasts. Its major advantage against univariate BMA is an increase in reliability when the forecast system is changing due to model availability.

Assessment of 21st century drought conditions at Shasta Dam based on dynamically projected water supply conditions by a regional climate model coupled with a physically-based hydrology model.

PubMed

Trinh, T; Ishida, K; Kavvas, M L; Ercan, A; Carr, K

2017-05-15

Along with socioeconomic developments, and population increase, natural disasters around the world have recently increased the awareness of harmful impacts they cause. Among natural disasters, drought is of great interest to scientists due to the extraordinary diversity of their severity and duration. Motivated by the development of a potential approach to investigate future possible droughts in a probabilistic framework based on climate change projections, a methodology to consider thirteen future climate projections based on four emission scenarios to characterize droughts is presented. The proposed approach uses a regional climate model coupled with a physically-based hydrology model (Watershed Environmental Hydrology Hydro-Climate Model; WEHY-HCM) to generate thirteen equally likely future water supply projections. The water supply projections were compared to the current water demand for the detection of drought events and estimation of drought properties. The procedure was applied to Shasta Dam watershed to analyze drought conditions at the watershed outlet, Shasta Dam. The results suggest an increasing water scarcity at Shasta Dam with more severe and longer future drought events in some future scenarios. An important advantage of the proposed approach to the probabilistic analysis of future droughts is that it provides the drought properties of the 100-year and 200-year return periods without resorting to any extrapolation of the frequency curve. Copyright © 2017 Elsevier B.V. All rights reserved.

Mapping flood hazards under uncertainty through probabilistic flood inundation maps

NASA Astrophysics Data System (ADS)

Stephens, T.; Bledsoe, B. P.; Miller, A. J.; Lee, G.

2017-12-01

Changing precipitation, rapid urbanization, and population growth interact to create unprecedented challenges for flood mitigation and management. Standard methods for estimating risk from flood inundation maps generally involve simulations of floodplain hydraulics for an established regulatory discharge of specified frequency. Hydraulic model results are then geospatially mapped and depicted as a discrete boundary of flood extents and a binary representation of the probability of inundation (in or out) that is assumed constant over a project's lifetime. Consequently, existing methods utilized to define flood hazards and assess risk management are hindered by deterministic approaches that assume stationarity in a nonstationary world, failing to account for spatio-temporal variability of climate and land use as they translate to hydraulic models. This presentation outlines novel techniques for portraying flood hazards and the results of multiple flood inundation maps spanning hydroclimatic regions. Flood inundation maps generated through modeling of floodplain hydraulics are probabilistic reflecting uncertainty quantified through Monte-Carlo analyses of model inputs and parameters under current and future scenarios. The likelihood of inundation and range of variability in flood extents resulting from Monte-Carlo simulations are then compared with deterministic evaluations of flood hazards from current regulatory flood hazard maps. By facilitating alternative approaches of portraying flood hazards, the novel techniques described in this presentation can contribute to a shifting paradigm in flood management that acknowledges the inherent uncertainty in model estimates and the nonstationary behavior of land use and climate.

Risk assessments of regional climate change over Europe: generation of probabilistic ensemble and uncertainty assessment for EURO-CODEX

NASA Astrophysics Data System (ADS)

Yuan, J.; Kopp, R. E.

2017-12-01

Quantitative risk analysis of regional climate change is crucial for risk management and impact assessment of climate change. Two major challenges to assessing the risks of climate change are: CMIP5 model runs, which drive EURO-CODEX downscaling runs, do not cover the full range of uncertainty of future projections; Climate models may underestimate the probability of tail risks (i.e. extreme events). To overcome the difficulties, this study offers a viable avenue, where a set of probabilistic climate ensemble is generated using the Surrogate/Model Mixed Ensemble (SMME) method. The probabilistic ensembles for temperature and precipitation are used to assess the range of uncertainty covered by five bias-corrected simulations from the high-resolution (0.11º) EURO-CODEX database, which are selected by the PESETA (The Projection of Economic impacts of climate change in Sectors of the European Union based on bottom-up Analysis) III project. Results show that the distribution of SMME ensemble is notably wider than both distribution of raw ensemble of GCMs and the spread of the five EURO-CORDEX in RCP8.5. Tail risks are well presented by the SMME ensemble. Both SMME ensemble and EURO-CORDEX projections are aggregated to administrative level, and are integrated into impact functions of PESETA III to assess climate risks in Europe. To further evaluate the uncertainties introduced by the downscaling process, we compare the 5 runs from EURO-CORDEX with runs from the corresponding GCMs. Time series of regional mean, spatial patterns, and climate indices are examined for the future climate (2080-2099) deviating from the present climate (1981-2010). The downscaling processes do not appear to be trend-preserving, e.g. the increase in regional mean temperature from EURO-CORDEX is slower than that from the corresponding GCM. The spatial pattern comparison reveals that the differences between each pair of GCM and EURO-CORDEX are small in winter. In summer, the temperatures of EURO-CORDEX are generally lower than those of GCMs, while the drying trends in precipitation of EURO-CORDEX are smaller than those of GCMs. Climate indices are significantly affected by bias-correction and downscaling process. Our study provides valuable information for selecting climate indices in different regions over Europe.

Classic articles and workbook: EPRI monographs on simulation of electric power production

DOE Office of Scientific and Technical Information (OSTI.GOV)

Stremel, J.P.

1991-12-01

This monograph republishes several articles including a seminal one on probabilistic production costing for electric power generation. That article is given in the original French along with a English translation. Another article, written by R. Booth, gives a popular explanation of the theory, and a workbook by B. Manhire is included that carries through a simple example step by step. The classical analysis of non-probabilistic generator dispatch by L.K. Kirchmayer is republished along with an introductory essay by J.P. Stremel that puts in perspective the monograph material. The article in French was written by H. Baleriaux, E. Jamoulle, and Fr.more » Linard de Guertechin and first published in Brussels in 1967. It derived a method for calculating the expected value of production costs by modifying a load duration curve through the use of probability factors that account for unplanned random generator outages. Although the paper showed how pump storage plants could be included and how linear programming could be applied, the convolution technique used in the probabilistic calculations is the part most widely applied. The tutorial paper by Booth was written in a light style, and its lucidity helped popularize the method. The workbook by Manhire also shows how the calculation can be shortened significantly using cumulants to approximate the load duration curve.« less

Information Retrieval Performance of Probabilistically Generated, Problem-Specific Computerized Provider Order Entry Pick-Lists: A Pilot Study

PubMed Central

Rothschild, Adam S.; Lehmann, Harold P.

2005-01-01

Objective: The aim of this study was to preliminarily determine the feasibility of probabilistically generating problem-specific computerized provider order entry (CPOE) pick-lists from a database of explicitly linked orders and problems from actual clinical cases. Design: In a pilot retrospective validation, physicians reviewed internal medicine cases consisting of the admission history and physical examination and orders placed using CPOE during the first 24 hours after admission. They created coded problem lists and linked orders from individual cases to the problem for which they were most indicated. Problem-specific order pick-lists were generated by including a given order in a pick-list if the probability of linkage of order and problem (PLOP) equaled or exceeded a specified threshold. PLOP for a given linked order-problem pair was computed as its prevalence among the other cases in the experiment with the given problem. The orders that the reviewer linked to a given problem instance served as the reference standard to evaluate its system-generated pick-list. Measurements: Recall, precision, and length of the pick-lists. Results: Average recall reached a maximum of .67 with a precision of .17 and pick-list length of 31.22 at a PLOP threshold of 0. Average precision reached a maximum of .73 with a recall of .09 and pick-list length of .42 at a PLOP threshold of .9. Recall varied inversely with precision in classic information retrieval behavior. Conclusion: We preliminarily conclude that it is feasible to generate problem-specific CPOE pick-lists probabilistically from a database of explicitly linked orders and problems. Further research is necessary to determine the usefulness of this approach in real-world settings. PMID:15684134

Composite Load Spectra for Select Space Propulsion Structural Components

NASA Technical Reports Server (NTRS)

Ho, Hing W.; Newell, James F.

1994-01-01

Generic load models are described with multiple levels of progressive sophistication to simulate the composite (combined) load spectra (CLS) that are induced in space propulsion system components, representative of Space Shuttle Main Engines (SSME), such as transfer ducts, turbine blades and liquid oxygen (LOX) posts. These generic (coupled) models combine the deterministic models for composite load dynamic, acoustic, high-pressure and high rotational speed, etc., load simulation using statistically varying coefficients. These coefficients are then determined using advanced probabilistic simulation methods with and without strategically selected experimental data. The entire simulation process is included in a CLS computer code. Applications of the computer code to various components in conjunction with the PSAM (Probabilistic Structural Analysis Method) to perform probabilistic load evaluation and life prediction evaluations are also described to illustrate the effectiveness of the coupled model approach.

On the accuracy and reproducibility of a novel probabilistic atlas-based generation for calculation of head attenuation maps on integrated PET/MR scanners.

PubMed

Chen, Kevin T; Izquierdo-Garcia, David; Poynton, Clare B; Chonde, Daniel B; Catana, Ciprian

2017-03-01

To propose an MR-based method for generating continuous-valued head attenuation maps and to assess its accuracy and reproducibility. Demonstrating that novel MR-based photon attenuation correction methods are both accurate and reproducible is essential prior to using them routinely in research and clinical studies on integrated PET/MR scanners. Continuous-valued linear attenuation coefficient maps ("μ-maps") were generated by combining atlases that provided the prior probability of voxel positions belonging to a certain tissue class (air, soft tissue, or bone) and an MR intensity-based likelihood classifier to produce posterior probability maps of tissue classes. These probabilities were used as weights to generate the μ-maps. The accuracy of this probabilistic atlas-based continuous-valued μ-map ("PAC-map") generation method was assessed by calculating the voxel-wise absolute relative change (RC) between the MR-based and scaled CT-based attenuation-corrected PET images. To assess reproducibility, we performed pair-wise comparisons of the RC values obtained from the PET images reconstructed using the μ-maps generated from the data acquired at three time points. The proposed method produced continuous-valued μ-maps that qualitatively reflected the variable anatomy in patients with brain tumor and agreed well with the scaled CT-based μ-maps. The absolute RC comparing the resulting PET volumes was 1.76 ± 2.33 %, quantitatively demonstrating that the method is accurate. Additionally, we also showed that the method is highly reproducible, the mean RC value for the PET images reconstructed using the μ-maps obtained at the three visits being 0.65 ± 0.95 %. Accurate and highly reproducible continuous-valued head μ-maps can be generated from MR data using a probabilistic atlas-based approach.

NetGen: a novel network-based probabilistic generative model for gene set functional enrichment analysis.

PubMed

Sun, Duanchen; Liu, Yinliang; Zhang, Xiang-Sun; Wu, Ling-Yun

2017-09-21

High-throughput experimental techniques have been dramatically improved and widely applied in the past decades. However, biological interpretation of the high-throughput experimental results, such as differential expression gene sets derived from microarray or RNA-seq experiments, is still a challenging task. Gene Ontology (GO) is commonly used in the functional enrichment studies. The GO terms identified via current functional enrichment analysis tools often contain direct parent or descendant terms in the GO hierarchical structure. Highly redundant terms make users difficult to analyze the underlying biological processes. In this paper, a novel network-based probabilistic generative model, NetGen, was proposed to perform the functional enrichment analysis. An additional protein-protein interaction (PPI) network was explicitly used to assist the identification of significantly enriched GO terms. NetGen achieved a superior performance than the existing methods in the simulation studies. The effectiveness of NetGen was explored further on four real datasets. Notably, several GO terms which were not directly linked with the active gene list for each disease were identified. These terms were closely related to the corresponding diseases when accessed to the curated literatures. NetGen has been implemented in the R package CopTea publicly available at GitHub ( http://github.com/wulingyun/CopTea/ ). Our procedure leads to a more reasonable and interpretable result of the functional enrichment analysis. As a novel term combination-based functional enrichment analysis method, NetGen is complementary to current individual term-based methods, and can help to explore the underlying pathogenesis of complex diseases.

Probabilistic modeling of the evolution of gene synteny within reconciled phylogenies

PubMed Central

2015-01-01

Background Most models of genome evolution concern either genetic sequences, gene content or gene order. They sometimes integrate two of the three levels, but rarely the three of them. Probabilistic models of gene order evolution usually have to assume constant gene content or adopt a presence/absence coding of gene neighborhoods which is blind to complex events modifying gene content. Results We propose a probabilistic evolutionary model for gene neighborhoods, allowing genes to be inserted, duplicated or lost. It uses reconciled phylogenies, which integrate sequence and gene content evolution. We are then able to optimize parameters such as phylogeny branch lengths, or probabilistic laws depicting the diversity of susceptibility of syntenic regions to rearrangements. We reconstruct a structure for ancestral genomes by optimizing a likelihood, keeping track of all evolutionary events at the level of gene content and gene synteny. Ancestral syntenies are associated with a probability of presence. We implemented the model with the restriction that at most one gene duplication separates two gene speciations in reconciled gene trees. We reconstruct ancestral syntenies on a set of 12 drosophila genomes, and compare the evolutionary rates along the branches and along the sites. We compare with a parsimony method and find a significant number of results not supported by the posterior probability. The model is implemented in the Bio++ library. It thus benefits from and enriches the classical models and methods for molecular evolution. PMID:26452018

A PROBABILISTIC EXPOSURE ASSESSMENT FOR CHILDREN WHO CONTACT CCA-TREATED PLAYSETS AND DECKS USING THE STOCHASTIC HUMAN EXPOSURE AND DOSE SIMULATION (SHEDS) MODEL FOR THE WOOD PRESERVATIVE EXPOSURE SCENARIO

EPA Science Inventory

The U.S. Environmental Protection Agency has conducted a probabilistic exposure and dose assessment on the arsenic (As) and chromium (Cr) components of Chromated Copper Arsenate (CCA) using the Stochastic Human Exposure and Dose Simulation model for wood preservatives (SHEDS-Wood...

Probabilistic Based Modeling and Simulation Assessment

DTIC Science & Technology

2010-06-01

different crash and blast scenarios. With the integration of the high fidelity neck and head model, a methodology to calculate the probability of injury...variability, correlation, and multiple (often competing) failure metrics. Important scenarios include vehicular collisions, blast /fragment impact, and...first area of focus is to develop a methodology to integrate probabilistic analysis into finite element analysis of vehicle collisions and blast . The

A Probabilistic Model for Students' Errors and Misconceptions on the Structure of Matter in Relation to Three Cognitive Variables

ERIC Educational Resources Information Center

Tsitsipis, Georgios; Stamovlasis, Dimitrios; Papageorgiou, George

2012-01-01

In this study, the effect of 3 cognitive variables such as logical thinking, field dependence/field independence, and convergent/divergent thinking on some specific students' answers related to the particulate nature of matter was investigated by means of probabilistic models. Besides recording and tabulating the students' responses, a combination…

Probabilistic commodity-flow-based focusing of monitoring activities to facilitate early detection of Phytophthora ramorum outbreaks

Treesearch

Steven C. McKelvey; William D. Smith; Frank Koch

2012-01-01

This project summary describes a probabilistic model developed with funding support from the Forest Health Monitoring Program of the Forest Service, U.S. Department of Agriculture (BaseEM Project SO-R-08-01). The model has been implemented in SODBuster, a standalone software package developed using the Java software development kit from Sun Microsystems.

Economic Dispatch for Microgrid Containing Electric Vehicles via Probabilistic Modeling: Preprint

DOE Office of Scientific and Technical Information (OSTI.GOV)

Yao, Yin; Gao, Wenzhong; Momoh, James

In this paper, an economic dispatch model with probabilistic modeling is developed for a microgrid. The electric power supply in a microgrid consists of conventional power plants and renewable energy power plants, such as wind and solar power plants. Because of the fluctuation in the output of solar and wind power plants, an empirical probabilistic model is developed to predict their hourly output. According to different characteristics of wind and solar power plants, the parameters for probabilistic distribution are further adjusted individually for both. On the other hand, with the growing trend in plug-in electric vehicles (PHEVs), an integrated microgridmore » system must also consider the impact of PHEVs. The charging loads from PHEVs as well as the discharging output via the vehicle-to-grid (V2G) method can greatly affect the economic dispatch for all of the micro energy sources in a microgrid. This paper presents an optimization method for economic dispatch in a microgrid considering conventional power plants, renewable power plants, and PHEVs. The simulation results reveal that PHEVs with V2G capability can be an indispensable supplement in a modern microgrid.« less

Probabilistic population projections with migration uncertainty

PubMed Central

Azose, Jonathan J.; Ševčíková, Hana; Raftery, Adrian E.

2016-01-01

We produce probabilistic projections of population for all countries based on probabilistic projections of fertility, mortality, and migration. We compare our projections to those from the United Nations’ Probabilistic Population Projections, which uses similar methods for fertility and mortality but deterministic migration projections. We find that uncertainty in migration projection is a substantial contributor to uncertainty in population projections for many countries. Prediction intervals for the populations of Northern America and Europe are over 70% wider, whereas prediction intervals for the populations of Africa, Asia, and the world as a whole are nearly unchanged. Out-of-sample validation shows that the model is reasonably well calibrated. PMID:27217571

Probabilistic brains: knowns and unknowns

PubMed Central

Pouget, Alexandre; Beck, Jeffrey M; Ma, Wei Ji; Latham, Peter E

2015-01-01

There is strong behavioral and physiological evidence that the brain both represents probability distributions and performs probabilistic inference. Computational neuroscientists have started to shed light on how these probabilistic representations and computations might be implemented in neural circuits. One particularly appealing aspect of these theories is their generality: they can be used to model a wide range of tasks, from sensory processing to high-level cognition. To date, however, these theories have only been applied to very simple tasks. Here we discuss the challenges that will emerge as researchers start focusing their efforts on real-life computations, with a focus on probabilistic learning, structural learning and approximate inference. PMID:23955561

A probabilistic asteroid impact risk model: assessment of sub-300 m impacts

NASA Astrophysics Data System (ADS)

Mathias, Donovan L.; Wheeler, Lorien F.; Dotson, Jessie L.

2017-06-01

A comprehensive asteroid threat assessment requires the quantification of both the impact likelihood and resulting consequence across the range of possible events. This paper presents a probabilistic asteroid impact risk (PAIR) assessment model developed for this purpose. The model incorporates published impact frequency rates with state-of-the-art consequence assessment tools, applied within a Monte Carlo framework that generates sets of impact scenarios from uncertain input parameter distributions. Explicit treatment of atmospheric entry is included to produce energy deposition rates that account for the effects of thermal ablation and object fragmentation. These energy deposition rates are used to model the resulting ground damage, and affected populations are computed for the sampled impact locations. The results for each scenario are aggregated into a distribution of potential outcomes that reflect the range of uncertain impact parameters, population densities, and strike probabilities. As an illustration of the utility of the PAIR model, the results are used to address the question of what minimum size asteroid constitutes a threat to the population. To answer this question, complete distributions of results are combined with a hypothetical risk tolerance posture to provide the minimum size, given sets of initial assumptions for objects up to 300 m in diameter. Model outputs demonstrate how such questions can be answered and provide a means for interpreting the effect that input assumptions and uncertainty can have on final risk-based decisions. Model results can be used to prioritize investments to gain knowledge in critical areas or, conversely, to identify areas where additional data have little effect on the metrics of interest.

Scope of Various Random Number Generators in Ant System Approach for TSP

NASA Technical Reports Server (NTRS)

Sen, S. K.; Shaykhian, Gholam Ali

2007-01-01

Experimented on heuristic, based on an ant system approach for traveling Salesman problem, are several quasi and pseudo-random number generators. This experiment is to explore if any particular generator is most desirable. Such an experiment on large samples has the potential to rank the performance of the generators for the foregoing heuristic. This is just to seek an answer to the controversial performance ranking of the generators in probabilistic/statically sense.

A probabilistic maintenance model for diesel engines

NASA Astrophysics Data System (ADS)

Pathirana, Shan; Abeygunawardane, Saranga Kumudu

2018-02-01

In this paper, a probabilistic maintenance model is developed for inspection based preventive maintenance of diesel engines based on the practical model concepts discussed in the literature. Developed model is solved using real data obtained from inspection and maintenance histories of diesel engines and experts' views. Reliability indices and costs were calculated for the present maintenance policy of diesel engines. A sensitivity analysis is conducted to observe the effect of inspection based preventive maintenance on the life cycle cost of diesel engines.

Using the Logarithm of Odds to Define a Vector Space on Probabilistic Atlases

PubMed Central

Pohl, Kilian M.; Fisher, John; Bouix, Sylvain; Shenton, Martha; McCarley, Robert W.; Grimson, W. Eric L.; Kikinis, Ron; Wells, William M.

2007-01-01

The Logarithm of the Odds ratio (LogOdds) is frequently used in areas such as artificial neural networks, economics, and biology, as an alternative representation of probabilities. Here, we use LogOdds to place probabilistic atlases in a linear vector space. This representation has several useful properties for medical imaging. For example, it not only encodes the shape of multiple anatomical structures but also captures some information concerning uncertainty. We demonstrate that the resulting vector space operations of addition and scalar multiplication have natural probabilistic interpretations. We discuss several examples for placing label maps into the space of LogOdds. First, we relate signed distance maps, a widely used implicit shape representation, to LogOdds and compare it to an alternative that is based on smoothing by spatial Gaussians. We find that the LogOdds approach better preserves shapes in a complex multiple object setting. In the second example, we capture the uncertainty of boundary locations by mapping multiple label maps of the same object into the LogOdds space. Third, we define a framework for non-convex interpolations among atlases that capture different time points in the aging process of a population. We evaluate the accuracy of our representation by generating a deformable shape atlas that captures the variations of anatomical shapes across a population. The deformable atlas is the result of a principal component analysis within the LogOdds space. This atlas is integrated into an existing segmentation approach for MR images. We compare the performance of the resulting implementation in segmenting 20 test cases to a similar approach that uses a more standard shape model that is based on signed distance maps. On this data set, the Bayesian classification model with our new representation outperformed the other approaches in segmenting subcortical structures. PMID:17698403

Size Evolution and Stochastic Models: Explaining Ostracod Size through Probabilistic Distributions

NASA Astrophysics Data System (ADS)

Krawczyk, M.; Decker, S.; Heim, N. A.; Payne, J.

2014-12-01

The biovolume of animals has functioned as an important benchmark for measuring evolution throughout geologic time. In our project, we examined the observed average body size of ostracods over time in order to understand the mechanism of size evolution in these marine organisms. The body size of ostracods has varied since the beginning of the Ordovician, where the first true ostracods appeared. We created a stochastic branching model to create possible evolutionary trees of ostracod size. Using stratigraphic ranges for ostracods compiled from over 750 genera in the Treatise on Invertebrate Paleontology, we calculated overall speciation and extinction rates for our model. At each timestep in our model, new lineages can evolve or existing lineages can become extinct. Newly evolved lineages are assigned sizes based on their parent genera. We parameterized our model to generate neutral and directional changes in ostracod size to compare with the observed data. New sizes were chosen via a normal distribution, and the neutral model selected new sizes differentials centered on zero, allowing for an equal chance of larger or smaller ostracods at each speciation. Conversely, the directional model centered the distribution on a negative value, giving a larger chance of smaller ostracods. Our data strongly suggests that the overall direction of ostracod evolution has been following a model that directionally pushes mean ostracod size down, shying away from a neutral model. Our model was able to match the magnitude of size decrease. Our models had a constant linear decrease while the actual data had a much more rapid initial rate followed by a constant size. The nuance of the observed trends ultimately suggests a more complex method of size evolution. In conclusion, probabilistic methods can provide valuable insight into possible evolutionary mechanisms determining size evolution in ostracods.

Multivariate Probabilistic Analysis of an Hydrological Model

NASA Astrophysics Data System (ADS)

Franceschini, Samuela; Marani, Marco

2010-05-01

Model predictions derived based on rainfall measurements and hydrological model results are often limited by the systematic error of measuring instruments, by the intrinsic variability of the natural processes and by the uncertainty of the mathematical representation. We propose a means to identify such sources of uncertainty and to quantify their effects based on point-estimate approaches, as a valid alternative to cumbersome Montecarlo methods. We present uncertainty analyses on the hydrologic response to selected meteorological events, in the mountain streamflow-generating portion of the Brenta basin at Bassano del Grappa, Italy. The Brenta river catchment has a relatively uniform morphology and quite a heterogeneous rainfall-pattern. In the present work, we evaluate two sources of uncertainty: data uncertainty (the uncertainty due to data handling and analysis) and model uncertainty (the uncertainty related to the formulation of the model). We thus evaluate the effects of the measurement error of tipping-bucket rain gauges, the uncertainty in estimating spatially-distributed rainfall through block kriging, and the uncertainty associated with estimated model parameters. To this end, we coupled a deterministic model based on the geomorphological theory of the hydrologic response to probabilistic methods. In particular we compare the results of Monte Carlo Simulations (MCS) to the results obtained, in the same conditions, using Li's Point Estimate Method (LiM). The LiM is a probabilistic technique that approximates the continuous probability distribution function of the considered stochastic variables by means of discrete points and associated weights. This allows to satisfactorily reproduce results with only few evaluations of the model function. The comparison between the LiM and MCS results highlights the pros and cons of using an approximating method. LiM is less computationally demanding than MCS, but has limited applicability especially when the model response is highly nonlinear. Higher-order approximations can provide more accurate estimations, but reduce the numerical advantage of the LiM. The results of the uncertainty analysis identify the main sources of uncertainty in the computation of river discharge. In this particular case the spatial variability of rainfall and the model parameters uncertainty are shown to have the greatest impact on discharge evaluation. This, in turn, highlights the need to support any estimated hydrological response with probability information and risk analysis results in order to provide a robust, systematic framework for decision making.

Probabilistic estimates of drought impacts on agricultural production

NASA Astrophysics Data System (ADS)

Madadgar, Shahrbanou; AghaKouchak, Amir; Farahmand, Alireza; Davis, Steven J.

2017-08-01

Increases in the severity and frequency of drought in a warming climate may negatively impact agricultural production and food security. Unlike previous studies that have estimated agricultural impacts of climate condition using single-crop yield distributions, we develop a multivariate probabilistic model that uses projected climatic conditions (e.g., precipitation amount or soil moisture) throughout a growing season to estimate the probability distribution of crop yields. We demonstrate the model by an analysis of the historical period 1980-2012, including the Millennium Drought in Australia (2001-2009). We find that precipitation and soil moisture deficit in dry growing seasons reduced the average annual yield of the five largest crops in Australia (wheat, broad beans, canola, lupine, and barley) by 25-45% relative to the wet growing seasons. Our model can thus produce region- and crop-specific agricultural sensitivities to climate conditions and variability. Probabilistic estimates of yield may help decision-makers in government and business to quantitatively assess the vulnerability of agriculture to climate variations. We develop a multivariate probabilistic model that uses precipitation to estimate the probability distribution of crop yields. The proposed model shows how the probability distribution of crop yield changes in response to droughts. During Australia's Millennium Drought precipitation and soil moisture deficit reduced the average annual yield of the five largest crops.

Probabilistic, meso-scale flood loss modelling

NASA Astrophysics Data System (ADS)

Kreibich, Heidi; Botto, Anna; Schröter, Kai; Merz, Bruno

2016-04-01

Flood risk analyses are an important basis for decisions on flood risk management and adaptation. However, such analyses are associated with significant uncertainty, even more if changes in risk due to global change are expected. Although uncertainty analysis and probabilistic approaches have received increased attention during the last years, they are still not standard practice for flood risk assessments and even more for flood loss modelling. State of the art in flood loss modelling is still the use of simple, deterministic approaches like stage-damage functions. Novel probabilistic, multi-variate flood loss models have been developed and validated on the micro-scale using a data-mining approach, namely bagging decision trees (Merz et al. 2013). In this presentation we demonstrate and evaluate the upscaling of the approach to the meso-scale, namely on the basis of land-use units. The model is applied in 19 municipalities which were affected during the 2002 flood by the River Mulde in Saxony, Germany (Botto et al. submitted). The application of bagging decision tree based loss models provide a probability distribution of estimated loss per municipality. Validation is undertaken on the one hand via a comparison with eight deterministic loss models including stage-damage functions as well as multi-variate models. On the other hand the results are compared with official loss data provided by the Saxon Relief Bank (SAB). The results show, that uncertainties of loss estimation remain high. Thus, the significant advantage of this probabilistic flood loss estimation approach is that it inherently provides quantitative information about the uncertainty of the prediction. References: Merz, B.; Kreibich, H.; Lall, U. (2013): Multi-variate flood damage assessment: a tree-based data-mining approach. NHESS, 13(1), 53-64. Botto A, Kreibich H, Merz B, Schröter K (submitted) Probabilistic, multi-variable flood loss modelling on the meso-scale with BT-FLEMO. Risk Analysis.

Probabilistic image modeling with an extended chain graph for human activity recognition and image segmentation.

PubMed

Zhang, Lei; Zeng, Zhi; Ji, Qiang

2011-09-01

Chain graph (CG) is a hybrid probabilistic graphical model (PGM) capable of modeling heterogeneous relationships among random variables. So far, however, its application in image and video analysis is very limited due to lack of principled learning and inference methods for a CG of general topology. To overcome this limitation, we introduce methods to extend the conventional chain-like CG model to CG model with more general topology and the associated methods for learning and inference in such a general CG model. Specifically, we propose techniques to systematically construct a generally structured CG, to parameterize this model, to derive its joint probability distribution, to perform joint parameter learning, and to perform probabilistic inference in this model. To demonstrate the utility of such an extended CG, we apply it to two challenging image and video analysis problems: human activity recognition and image segmentation. The experimental results show improved performance of the extended CG model over the conventional directed or undirected PGMs. This study demonstrates the promise of the extended CG for effective modeling and inference of complex real-world problems.