Advanced obstacle avoidance for a laser based wheelchair using optimised Bayesian neural networks.

PubMed

Trieu, Hoang T; Nguyen, Hung T; Willey, Keith

2008-01-01

In this paper we present an advanced method of obstacle avoidance for a laser based intelligent wheelchair using optimized Bayesian neural networks. Three neural networks are designed for three separate sub-tasks: passing through a door way, corridor and wall following and general obstacle avoidance. The accurate usable accessible space is determined by including the actual wheelchair dimensions in a real-time map used as inputs to each networks. Data acquisitions are performed separately to collect the patterns required for specified sub-tasks. Bayesian frame work is used to determine the optimal neural network structure in each case. Then these networks are trained under the supervision of Bayesian rule. Experiment results showed that compare to the VFH algorithm our neural networks navigated a smoother path following a near optimum trajectory.

Machine Learning Topological Invariants with Neural Networks

NASA Astrophysics Data System (ADS)

Zhang, Pengfei; Shen, Huitao; Zhai, Hui

2018-02-01

In this Letter we supervisedly train neural networks to distinguish different topological phases in the context of topological band insulators. After training with Hamiltonians of one-dimensional insulators with chiral symmetry, the neural network can predict their topological winding numbers with nearly 100% accuracy, even for Hamiltonians with larger winding numbers that are not included in the training data. These results show a remarkable success that the neural network can capture the global and nonlinear topological features of quantum phases from local inputs. By opening up the neural network, we confirm that the network does learn the discrete version of the winding number formula. We also make a couple of remarks regarding the role of the symmetry and the opposite effect of regularization techniques when applying machine learning to physical systems.

Receptive field optimisation and supervision of a fuzzy spiking neural network.

PubMed

Glackin, Cornelius; Maguire, Liam; McDaid, Liam; Sayers, Heather

2011-04-01

This paper presents a supervised training algorithm that implements fuzzy reasoning on a spiking neural network. Neuron selectivity is facilitated using receptive fields that enable individual neurons to be responsive to certain spike train firing rates and behave in a similar manner as fuzzy membership functions. The connectivity of the hidden and output layers in the fuzzy spiking neural network (FSNN) is representative of a fuzzy rule base. Fuzzy C-Means clustering is utilised to produce clusters that represent the antecedent part of the fuzzy rule base that aid classification of the feature data. Suitable cluster widths are determined using two strategies; subjective thresholding and evolutionary thresholding respectively. The former technique typically results in compact solutions in terms of the number of neurons, and is shown to be particularly suited to small data sets. In the latter technique a pool of cluster candidates is generated using Fuzzy C-Means clustering and then a genetic algorithm is employed to select the most suitable clusters and to specify cluster widths. In both scenarios, the network is supervised but learning only occurs locally as in the biological case. The advantages and disadvantages of the network topology for the Fisher Iris and Wisconsin Breast Cancer benchmark classification tasks are demonstrated and directions of current and future work are discussed. Copyright © 2010 Elsevier Ltd. All rights reserved.

Reducing weight precision of convolutional neural networks towards large-scale on-chip image recognition

NASA Astrophysics Data System (ADS)

Ji, Zhengping; Ovsiannikov, Ilia; Wang, Yibing; Shi, Lilong; Zhang, Qiang

2015-05-01

In this paper, we develop a server-client quantization scheme to reduce bit resolution of deep learning architecture, i.e., Convolutional Neural Networks, for image recognition tasks. Low bit resolution is an important factor in bringing the deep learning neural network into hardware implementation, which directly determines the cost and power consumption. We aim to reduce the bit resolution of the network without sacrificing its performance. To this end, we design a new quantization algorithm called supervised iterative quantization to reduce the bit resolution of learned network weights. In the training stage, the supervised iterative quantization is conducted via two steps on server - apply k-means based adaptive quantization on learned network weights and retrain the network based on quantized weights. These two steps are alternated until the convergence criterion is met. In this testing stage, the network configuration and low-bit weights are loaded to the client hardware device to recognize coming input in real time, where optimized but expensive quantization becomes infeasible. Considering this, we adopt a uniform quantization for the inputs and internal network responses (called feature maps) to maintain low on-chip expenses. The Convolutional Neural Network with reduced weight and input/response precision is demonstrated in recognizing two types of images: one is hand-written digit images and the other is real-life images in office scenarios. Both results show that the new network is able to achieve the performance of the neural network with full bit resolution, even though in the new network the bit resolution of both weight and input are significantly reduced, e.g., from 64 bits to 4-5 bits.

Classifying galaxy spectra at 0.5 < z < 1 with self-organizing maps

NASA Astrophysics Data System (ADS)

Rahmani, S.; Teimoorinia, H.; Barmby, P.

2018-05-01

The spectrum of a galaxy contains information about its physical properties. Classifying spectra using templates helps elucidate the nature of a galaxy's energy sources. In this paper, we investigate the use of self-organizing maps in classifying galaxy spectra against templates. We trained semi-supervised self-organizing map networks using a set of templates covering the wavelength range from far ultraviolet to near infrared. The trained networks were used to classify the spectra of a sample of 142 galaxies with 0.5 < z < 1 and the results compared to classifications performed using K-means clustering, a supervised neural network, and chi-squared minimization. Spectra corresponding to quiescent galaxies were more likely to be classified similarly by all methods while starburst spectra showed more variability. Compared to classification using chi-squared minimization or the supervised neural network, the galaxies classed together by the self-organizing map had more similar spectra. The class ordering provided by the one-dimensional self-organizing maps corresponds to an ordering in physical properties, a potentially important feature for the exploration of large datasets.

Supervised Machine Learning for Regionalization of Environmental Data: Distribution of Uranium in Groundwater in Ukraine

NASA Astrophysics Data System (ADS)

Govorov, Michael; Gienko, Gennady; Putrenko, Viktor

2018-05-01

In this paper, several supervised machine learning algorithms were explored to define homogeneous regions of con-centration of uranium in surface waters in Ukraine using multiple environmental parameters. The previous study was focused on finding the primary environmental parameters related to uranium in ground waters using several methods of spatial statistics and unsupervised classification. At this step, we refined the regionalization using Artifi-cial Neural Networks (ANN) techniques including Multilayer Perceptron (MLP), Radial Basis Function (RBF), and Convolutional Neural Network (CNN). The study is focused on building local ANN models which may significantly improve the prediction results of machine learning algorithms by taking into considerations non-stationarity and autocorrelation in spatial data.

A comparison of neural network and fuzzy clustering techniques in segmenting magnetic resonance images of the brain.

PubMed

Hall, L O; Bensaid, A M; Clarke, L P; Velthuizen, R P; Silbiger, M S; Bezdek, J C

1992-01-01

Magnetic resonance (MR) brain section images are segmented and then synthetically colored to give visual representations of the original data with three approaches: the literal and approximate fuzzy c-means unsupervised clustering algorithms, and a supervised computational neural network. Initial clinical results are presented on normal volunteers and selected patients with brain tumors surrounded by edema. Supervised and unsupervised segmentation techniques provide broadly similar results. Unsupervised fuzzy algorithms were visually observed to show better segmentation when compared with raw image data for volunteer studies. For a more complex segmentation problem with tumor/edema or cerebrospinal fluid boundary, where the tissues have similar MR relaxation behavior, inconsistency in rating among experts was observed, with fuzz-c-means approaches being slightly preferred over feedforward cascade correlation results. Various facets of both approaches, such as supervised versus unsupervised learning, time complexity, and utility for the diagnostic process, are compared.

Automatic Classification Using Supervised Learning in a Medical Document Filtering Application.

ERIC Educational Resources Information Center

Mostafa, J.; Lam, W.

2000-01-01

Presents a multilevel model of the information filtering process that permits document classification. Evaluates a document classification approach based on a supervised learning algorithm, measures the accuracy of the algorithm in a neural network that was trained to classify medical documents on cell biology, and discusses filtering…

Neural network with dynamically adaptable neurons

NASA Technical Reports Server (NTRS)

Tawel, Raoul (Inventor)

1994-01-01

This invention is an adaptive neuron for use in neural network processors. The adaptive neuron participates in the supervised learning phase of operation on a co-equal basis with the synapse matrix elements by adaptively changing its gain in a similar manner to the change of weights in the synapse IO elements. In this manner, training time is decreased by as much as three orders of magnitude.

The Role of Architectural and Learning Constraints in Neural Network Models: A Case Study on Visual Space Coding.

PubMed

Testolin, Alberto; De Filippo De Grazia, Michele; Zorzi, Marco

2017-01-01

The recent "deep learning revolution" in artificial neural networks had strong impact and widespread deployment for engineering applications, but the use of deep learning for neurocomputational modeling has been so far limited. In this article we argue that unsupervised deep learning represents an important step forward for improving neurocomputational models of perception and cognition, because it emphasizes the role of generative learning as opposed to discriminative (supervised) learning. As a case study, we present a series of simulations investigating the emergence of neural coding of visual space for sensorimotor transformations. We compare different network architectures commonly used as building blocks for unsupervised deep learning by systematically testing the type of receptive fields and gain modulation developed by the hidden neurons. In particular, we compare Restricted Boltzmann Machines (RBMs), which are stochastic, generative networks with bidirectional connections trained using contrastive divergence, with autoencoders, which are deterministic networks trained using error backpropagation. For both learning architectures we also explore the role of sparse coding, which has been identified as a fundamental principle of neural computation. The unsupervised models are then compared with supervised, feed-forward networks that learn an explicit mapping between different spatial reference frames. Our simulations show that both architectural and learning constraints strongly influenced the emergent coding of visual space in terms of distribution of tuning functions at the level of single neurons. Unsupervised models, and particularly RBMs, were found to more closely adhere to neurophysiological data from single-cell recordings in the primate parietal cortex. These results provide new insights into how basic properties of artificial neural networks might be relevant for modeling neural information processing in biological systems.

The Role of Architectural and Learning Constraints in Neural Network Models: A Case Study on Visual Space Coding

PubMed Central

Testolin, Alberto; De Filippo De Grazia, Michele; Zorzi, Marco

2017-01-01

The recent “deep learning revolution” in artificial neural networks had strong impact and widespread deployment for engineering applications, but the use of deep learning for neurocomputational modeling has been so far limited. In this article we argue that unsupervised deep learning represents an important step forward for improving neurocomputational models of perception and cognition, because it emphasizes the role of generative learning as opposed to discriminative (supervised) learning. As a case study, we present a series of simulations investigating the emergence of neural coding of visual space for sensorimotor transformations. We compare different network architectures commonly used as building blocks for unsupervised deep learning by systematically testing the type of receptive fields and gain modulation developed by the hidden neurons. In particular, we compare Restricted Boltzmann Machines (RBMs), which are stochastic, generative networks with bidirectional connections trained using contrastive divergence, with autoencoders, which are deterministic networks trained using error backpropagation. For both learning architectures we also explore the role of sparse coding, which has been identified as a fundamental principle of neural computation. The unsupervised models are then compared with supervised, feed-forward networks that learn an explicit mapping between different spatial reference frames. Our simulations show that both architectural and learning constraints strongly influenced the emergent coding of visual space in terms of distribution of tuning functions at the level of single neurons. Unsupervised models, and particularly RBMs, were found to more closely adhere to neurophysiological data from single-cell recordings in the primate parietal cortex. These results provide new insights into how basic properties of artificial neural networks might be relevant for modeling neural information processing in biological systems. PMID:28377709

Fuzzy-neural control of an aircraft tracking camera platform

NASA Technical Reports Server (NTRS)

Mcgrath, Dennis

1994-01-01

A fuzzy-neural control system simulation was developed for the control of a camera platform used to observe aircraft on final approach to an aircraft carrier. The fuzzy-neural approach to control combines the structure of a fuzzy knowledge base with a supervised neural network's ability to adapt and improve. The performance characteristics of this hybrid system were compared to those of a fuzzy system and a neural network system developed independently to determine if the fusion of these two technologies offers any advantage over the use of one or the other. The results of this study indicate that the fuzzy-neural approach to control offers some advantages over either fuzzy or neural control alone.

Neural networks for data mining electronic text collections

NASA Astrophysics Data System (ADS)

Walker, Nicholas; Truman, Gregory

1997-04-01

The use of neural networks in information retrieval and text analysis has primarily suffered from the issues of adequate document representation, the ability to scale to very large collections, dynamism in the face of new information and the practical difficulties of basing the design on the use of supervised training sets. Perhaps the most important approach to begin solving these problems is the use of `intermediate entities' which reduce the dimensionality of document representations and the size of documents collections to manageable levels coupled with the use of unsupervised neural network paradigms. This paper describes the issues, a fully configured neural network-based text analysis system--dataHARVEST--aimed at data mining text collections which begins this process, along with the remaining difficulties and potential ways forward.

Broad Absorption Line Quasar catalogues with Supervised Neural Networks

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

Scaringi, Simone; Knigge, Christian; Cottis, Christopher E.

2008-12-05

We have applied a Learning Vector Quantization (LVQ) algorithm to SDSS DR5 quasar spectra in order to create a large catalogue of broad absorption line quasars (BALQSOs). We first discuss the problems with BALQSO catalogues constructed using the conventional balnicity and/or absorption indices (BI and AI), and then describe the supervised LVQ network we have trained to recognise BALQSOs. The resulting BALQSO catalogue should be substantially more robust and complete than BI-or AI-based ones.

Spiking neural networks for handwritten digit recognition-Supervised learning and network optimization.

PubMed

Kulkarni, Shruti R; Rajendran, Bipin

2018-07-01

We demonstrate supervised learning in Spiking Neural Networks (SNNs) for the problem of handwritten digit recognition using the spike triggered Normalized Approximate Descent (NormAD) algorithm. Our network that employs neurons operating at sparse biological spike rates below 300Hz achieves a classification accuracy of 98.17% on the MNIST test database with four times fewer parameters compared to the state-of-the-art. We present several insights from extensive numerical experiments regarding optimization of learning parameters and network configuration to improve its accuracy. We also describe a number of strategies to optimize the SNN for implementation in memory and energy constrained hardware, including approximations in computing the neuronal dynamics and reduced precision in storing the synaptic weights. Experiments reveal that even with 3-bit synaptic weights, the classification accuracy of the designed SNN does not degrade beyond 1% as compared to the floating-point baseline. Further, the proposed SNN, which is trained based on the precise spike timing information outperforms an equivalent non-spiking artificial neural network (ANN) trained using back propagation, especially at low bit precision. Thus, our study shows the potential for realizing efficient neuromorphic systems that use spike based information encoding and learning for real-world applications. Copyright © 2018 Elsevier Ltd. All rights reserved.

Spatial features of synaptic adaptation affecting learning performance.

PubMed

Berger, Damian L; de Arcangelis, Lucilla; Herrmann, Hans J

2017-09-08

Recent studies have proposed that the diffusion of messenger molecules, such as monoamines, can mediate the plastic adaptation of synapses in supervised learning of neural networks. Based on these findings we developed a model for neural learning, where the signal for plastic adaptation is assumed to propagate through the extracellular space. We investigate the conditions allowing learning of Boolean rules in a neural network. Even fully excitatory networks show very good learning performances. Moreover, the investigation of the plastic adaptation features optimizing the performance suggests that learning is very sensitive to the extent of the plastic adaptation and the spatial range of synaptic connections.

CNN: a speaker recognition system using a cascaded neural network.

PubMed

Zaki, M; Ghalwash, A; Elkouny, A A

1996-05-01

The main emphasis of this paper is to present an approach for combining supervised and unsupervised neural network models to the issue of speaker recognition. To enhance the overall operation and performance of recognition, the proposed strategy integrates the two techniques, forming one global model called the cascaded model. We first present a simple conventional technique based on the distance measured between a test vector and a reference vector for different speakers in the population. This particular distance metric has the property of weighting down the components in those directions along which the intraspeaker variance is large. The reason for presenting this method is to clarify the discrepancy in performance between the conventional and neural network approach. We then introduce the idea of using unsupervised learning technique, presented by the winner-take-all model, as a means of recognition. Due to several tests that have been conducted and in order to enhance the performance of this model, dealing with noisy patterns, we have preceded it with a supervised learning model--the pattern association model--which acts as a filtration stage. This work includes both the design and implementation of both conventional and neural network approaches to recognize the speakers templates--which are introduced to the system via a voice master card and preprocessed before extracting the features used in the recognition. The conclusion indicates that the system performance in case of neural network is better than that of the conventional one, achieving a smooth degradation in respect of noisy patterns, and higher performance in respect of noise-free patterns.

Protein secondary structure prediction using modular reciprocal bidirectional recurrent neural networks.

PubMed

Babaei, Sepideh; Geranmayeh, Amir; Seyyedsalehi, Seyyed Ali

2010-12-01

The supervised learning of recurrent neural networks well-suited for prediction of protein secondary structures from the underlying amino acids sequence is studied. Modular reciprocal recurrent neural networks (MRR-NN) are proposed to model the strong correlations between adjacent secondary structure elements. Besides, a multilayer bidirectional recurrent neural network (MBR-NN) is introduced to capture the long-range intramolecular interactions between amino acids in formation of the secondary structure. The final modular prediction system is devised based on the interactive integration of the MRR-NN and the MBR-NN structures to arbitrarily engage the neighboring effects of the secondary structure types concurrent with memorizing the sequential dependencies of amino acids along the protein chain. The advanced combined network augments the percentage accuracy (Q₃) to 79.36% and boosts the segment overlap (SOV) up to 70.09% when tested on the PSIPRED dataset in three-fold cross-validation. Copyright © 2010 Elsevier Ireland Ltd. All rights reserved.

Rapid analysis of microbial systems using vibrational spectroscopy and supervised learning methods: application to the discrimination between methicillin-resistant and methicillin-susceptible Staphy

NASA Astrophysics Data System (ADS)

Goodacre, Royston; Rooney, Paul J.; Kell, Douglas B.

1998-04-01

FTIR spectra were obtained from 15 methicillin-resistant and 22 methicillin-susceptible Staphylococcus aureus strains using our DRASTIC approach. Cluster analysis showed that the major source of variation between the IR spectra was not due to their resistance or susceptibility to methicillin; indeed early studies suing pyrolysis mass spectrometry had shown that this unsupervised analysis gave information on the phage group of the bacteria. By contrast, artificial neural networks, based on a supervised learning, could be trained to recognize those aspects of the IR spectra which differentiated methicillin-resistant from methicillin- susceptible strains. These results give the first demonstration that the combination of FTIR with neural networks can provide a very rapid and accurate antibiotic susceptibility testing technique.

A comparison of neural network and fuzzy clustering techniques in segmenting magnetic resonance images of the brain

NASA Technical Reports Server (NTRS)

Hall, Lawrence O.; Bensaid, Amine M.; Clarke, Laurence P.; Velthuizen, Robert P.; Silbiger, Martin S.; Bezdek, James C.

1992-01-01

Magnetic resonance (MR) brain section images are segmented and then synthetically colored to give visual representations of the original data with three approaches: the literal and approximate fuzzy c-means unsupervised clustering algorithms and a supervised computational neural network, a dynamic multilayered perception trained with the cascade correlation learning algorithm. Initial clinical results are presented on both normal volunteers and selected patients with brain tumors surrounded by edema. Supervised and unsupervised segmentation techniques provide broadly similar results. Unsupervised fuzzy algorithms were visually observed to show better segmentation when compared with raw image data for volunteer studies. However, for a more complex segmentation problem with tumor/edema or cerebrospinal fluid boundary, where the tissues have similar MR relaxation behavior, inconsistency in rating among experts was observed.

Neural Network Classifies Teleoperation Data

NASA Technical Reports Server (NTRS)

Fiorini, Paolo; Giancaspro, Antonio; Losito, Sergio; Pasquariello, Guido

1994-01-01

Prototype artificial neural network, implemented in software, identifies phases of telemanipulator tasks in real time by analyzing feedback signals from force sensors on manipulator hand. Prototype is early, subsystem-level product of continuing effort to develop automated system that assists in training and supervising human control operator: provides symbolic feedback (e.g., warnings of impending collisions or evaluations of performance) to operator in real time during successive executions of same task. Also simplifies transition between teleoperation and autonomous modes of telerobotic system.

A neural network with modular hierarchical learning

NASA Technical Reports Server (NTRS)

Baldi, Pierre F. (Inventor); Toomarian, Nikzad (Inventor)

1994-01-01

This invention provides a new hierarchical approach for supervised neural learning of time dependent trajectories. The modular hierarchical methodology leads to architectures which are more structured than fully interconnected networks. The networks utilize a general feedforward flow of information and sparse recurrent connections to achieve dynamic effects. The advantages include the sparsity of units and connections, the modular organization. A further advantage is that the learning is much more circumscribed learning than in fully interconnected systems. The present invention is embodied by a neural network including a plurality of neural modules each having a pre-established performance capability wherein each neural module has an output outputting present results of the performance capability and an input for changing the present results of the performance capabilitiy. For pattern recognition applications, the performance capability may be an oscillation capability producing a repeating wave pattern as the present results. In the preferred embodiment, each of the plurality of neural modules includes a pre-established capability portion and a performance adjustment portion connected to control the pre-established capability portion.

Reward-based training of recurrent neural networks for cognitive and value-based tasks

PubMed Central

Song, H Francis; Yang, Guangyu R; Wang, Xiao-Jing

2017-01-01

Trained neural network models, which exhibit features of neural activity recorded from behaving animals, may provide insights into the circuit mechanisms of cognitive functions through systematic analysis of network activity and connectivity. However, in contrast to the graded error signals commonly used to train networks through supervised learning, animals learn from reward feedback on definite actions through reinforcement learning. Reward maximization is particularly relevant when optimal behavior depends on an animal’s internal judgment of confidence or subjective preferences. Here, we implement reward-based training of recurrent neural networks in which a value network guides learning by using the activity of the decision network to predict future reward. We show that such models capture behavioral and electrophysiological findings from well-known experimental paradigms. Our work provides a unified framework for investigating diverse cognitive and value-based computations, and predicts a role for value representation that is essential for learning, but not executing, a task. DOI: http://dx.doi.org/10.7554/eLife.21492.001 PMID:28084991

Semi-supervised anomaly detection - towards model-independent searches of new physics

NASA Astrophysics Data System (ADS)

Kuusela, Mikael; Vatanen, Tommi; Malmi, Eric; Raiko, Tapani; Aaltonen, Timo; Nagai, Yoshikazu

2012-06-01

Most classification algorithms used in high energy physics fall under the category of supervised machine learning. Such methods require a training set containing both signal and background events and are prone to classification errors should this training data be systematically inaccurate for example due to the assumed MC model. To complement such model-dependent searches, we propose an algorithm based on semi-supervised anomaly detection techniques, which does not require a MC training sample for the signal data. We first model the background using a multivariate Gaussian mixture model. We then search for deviations from this model by fitting to the observations a mixture of the background model and a number of additional Gaussians. This allows us to perform pattern recognition of any anomalous excess over the background. We show by a comparison to neural network classifiers that such an approach is a lot more robust against misspecification of the signal MC than supervised classification. In cases where there is an unexpected signal, a neural network might fail to correctly identify it, while anomaly detection does not suffer from such a limitation. On the other hand, when there are no systematic errors in the training data, both methods perform comparably.

Neural networks for learning and prediction with applications to remote sensing and speech perception

NASA Astrophysics Data System (ADS)

Gjaja, Marin N.

1997-11-01

Neural networks for supervised and unsupervised learning are developed and applied to problems in remote sensing, continuous map learning, and speech perception. Adaptive Resonance Theory (ART) models are real-time neural networks for category learning, pattern recognition, and prediction. Unsupervised fuzzy ART networks synthesize fuzzy logic and neural networks, and supervised ARTMAP networks incorporate ART modules for prediction and classification. New ART and ARTMAP methods resulting from analyses of data structure, parameter specification, and category selection are developed. Architectural modifications providing flexibility for a variety of applications are also introduced and explored. A new methodology for automatic mapping from Landsat Thematic Mapper (TM) and terrain data, based on fuzzy ARTMAP, is developed. System capabilities are tested on a challenging remote sensing problem, prediction of vegetation classes in the Cleveland National Forest from spectral and terrain features. After training at the pixel level, performance is tested at the stand level, using sites not seen during training. Results are compared to those of maximum likelihood classifiers, back propagation neural networks, and K-nearest neighbor algorithms. Best performance is obtained using a hybrid system based on a convex combination of fuzzy ARTMAP and maximum likelihood predictions. This work forms the foundation for additional studies exploring fuzzy ARTMAP's capability to estimate class mixture composition for non-homogeneous sites. Exploratory simulations apply ARTMAP to the problem of learning continuous multidimensional mappings. A novel system architecture retains basic ARTMAP properties of incremental and fast learning in an on-line setting while adding components to solve this class of problems. The perceptual magnet effect is a language-specific phenomenon arising early in infant speech development that is characterized by a warping of speech sound perception. An unsupervised neural network model is proposed that embodies two principal hypotheses supported by experimental data--that sensory experience guides language-specific development of an auditory neural map and that a population vector can predict psychological phenomena based on map cell activities. Model simulations show how a nonuniform distribution of map cell firing preferences can develop from language-specific input and give rise to the magnet effect.

The graph neural network model.

PubMed

Scarselli, Franco; Gori, Marco; Tsoi, Ah Chung; Hagenbuchner, Markus; Monfardini, Gabriele

2009-01-01

Many underlying relationships among data in several areas of science and engineering, e.g., computer vision, molecular chemistry, molecular biology, pattern recognition, and data mining, can be represented in terms of graphs. In this paper, we propose a new neural network model, called graph neural network (GNN) model, that extends existing neural network methods for processing the data represented in graph domains. This GNN model, which can directly process most of the practically useful types of graphs, e.g., acyclic, cyclic, directed, and undirected, implements a function tau(G,n) is an element of IR(m) that maps a graph G and one of its nodes n into an m-dimensional Euclidean space. A supervised learning algorithm is derived to estimate the parameters of the proposed GNN model. The computational cost of the proposed algorithm is also considered. Some experimental results are shown to validate the proposed learning algorithm, and to demonstrate its generalization capabilities.

Two States Mapping Based Time Series Neural Network Model for Compensation Prediction Residual Error

NASA Astrophysics Data System (ADS)

Jung, Insung; Koo, Lockjo; Wang, Gi-Nam

2008-11-01

The objective of this paper was to design a model of human bio signal data prediction system for decreasing of prediction error using two states mapping based time series neural network BP (back-propagation) model. Normally, a lot of the industry has been applied neural network model by training them in a supervised manner with the error back-propagation algorithm for time series prediction systems. However, it still has got a residual error between real value and prediction result. Therefore, we designed two states of neural network model for compensation residual error which is possible to use in the prevention of sudden death and metabolic syndrome disease such as hypertension disease and obesity. We determined that most of the simulation cases were satisfied by the two states mapping based time series prediction model. In particular, small sample size of times series were more accurate than the standard MLP model.

RM-SORN: a reward-modulated self-organizing recurrent neural network.

PubMed

Aswolinskiy, Witali; Pipa, Gordon

2015-01-01

Neural plasticity plays an important role in learning and memory. Reward-modulation of plasticity offers an explanation for the ability of the brain to adapt its neural activity to achieve a rewarded goal. Here, we define a neural network model that learns through the interaction of Intrinsic Plasticity (IP) and reward-modulated Spike-Timing-Dependent Plasticity (STDP). IP enables the network to explore possible output sequences and STDP, modulated by reward, reinforces the creation of the rewarded output sequences. The model is tested on tasks for prediction, recall, non-linear computation, pattern recognition, and sequence generation. It achieves performance comparable to networks trained with supervised learning, while using simple, biologically motivated plasticity rules, and rewarding strategies. The results confirm the importance of investigating the interaction of several plasticity rules in the context of reward-modulated learning and whether reward-modulated self-organization can explain the amazing capabilities of the brain.

Using Artificial Neural Networks to Predict the Presence of Overpressured Zones in the Anadarko Basin, Oklahoma

NASA Astrophysics Data System (ADS)

Cranganu, Constantin

2007-10-01

Many sedimentary basins throughout the world exhibit areas with abnormal pore-fluid pressures (higher or lower than normal or hydrostatic pressure). Predicting pore pressure and other parameters (depth, extension, magnitude, etc.) in such areas are challenging tasks. The compressional acoustic (sonic) log (DT) is often used as a predictor because it responds to changes in porosity or compaction produced by abnormal pore-fluid pressures. Unfortunately, the sonic log is not commonly recorded in most oil and/or gas wells. We propose using an artificial neural network to synthesize sonic logs by identifying the mathematical dependency between DT and the commonly available logs, such as normalized gamma ray (GR) and deep resistivity logs (REID). The artificial neural network process can be divided into three steps: (1) Supervised training of the neural network; (2) confirmation and validation of the model by blind-testing the results in wells that contain both the predictor (GR, REID) and the target values (DT) used in the supervised training; and 3) applying the predictive model to all wells containing the required predictor data and verifying the accuracy of the synthetic DT data by comparing the back-predicted synthetic predictor curves (GRNN, REIDNN) to the recorded predictor curves used in training (GR, REID). Artificial neural networks offer significant advantages over traditional deterministic methods. They do not require a precise mathematical model equation that describes the dependency between the predictor values and the target values and, unlike linear regression techniques, neural network methods do not overpredict mean values and thereby preserve original data variability. One of their most important advantages is that their predictions can be validated and confirmed through back-prediction of the input data. This procedure was applied to predict the presence of overpressured zones in the Anadarko Basin, Oklahoma. The results are promising and encouraging.

Modeling and control of magnetorheological fluid dampers using neural networks

NASA Astrophysics Data System (ADS)

Wang, D. H.; Liao, W. H.

2005-02-01

Due to the inherent nonlinear nature of magnetorheological (MR) fluid dampers, one of the challenging aspects for utilizing these devices to achieve high system performance is the development of accurate models and control algorithms that can take advantage of their unique characteristics. In this paper, the direct identification and inverse dynamic modeling for MR fluid dampers using feedforward and recurrent neural networks are studied. The trained direct identification neural network model can be used to predict the damping force of the MR fluid damper on line, on the basis of the dynamic responses across the MR fluid damper and the command voltage, and the inverse dynamic neural network model can be used to generate the command voltage according to the desired damping force through supervised learning. The architectures and the learning methods of the dynamic neural network models and inverse neural network models for MR fluid dampers are presented, and some simulation results are discussed. Finally, the trained neural network models are applied to predict and control the damping force of the MR fluid damper. Moreover, validation methods for the neural network models developed are proposed and used to evaluate their performance. Validation results with different data sets indicate that the proposed direct identification dynamic model using the recurrent neural network can be used to predict the damping force accurately and the inverse identification dynamic model using the recurrent neural network can act as a damper controller to generate the command voltage when the MR fluid damper is used in a semi-active mode.

Supervised neural network classification of pre-sliced cooked pork ham images using quaternionic singular values.

PubMed

Valous, Nektarios A; Mendoza, Fernando; Sun, Da-Wen; Allen, Paul

2010-03-01

The quaternionic singular value decomposition is a technique to decompose a quaternion matrix (representation of a colour image) into quaternion singular vector and singular value component matrices exposing useful properties. The objective of this study was to use a small portion of uncorrelated singular values, as robust features for the classification of sliced pork ham images, using a supervised artificial neural network classifier. Images were acquired from four qualities of sliced cooked pork ham typically consumed in Ireland (90 slices per quality), having similar appearances. Mahalanobis distances and Pearson product moment correlations were used for feature selection. Six highly discriminating features were used as input to train the neural network. An adaptive feedforward multilayer perceptron classifier was employed to obtain a suitable mapping from the input dataset. The overall correct classification performance for the training, validation and test set were 90.3%, 94.4%, and 86.1%, respectively. The results confirm that the classification performance was satisfactory. Extracting the most informative features led to the recognition of a set of different but visually quite similar textural patterns based on quaternionic singular values. Copyright 2009 Elsevier Ltd. All rights reserved.

Embeddable Reconfigurable Neuroprocessors

NASA Technical Reports Server (NTRS)

Daud, Taher; Duong, Tuan; Langenbacher, Harry; Tran, Mua; Thakoor, Anil

1993-01-01

Reconfigurable and cascadable building block neural network chips, fabricated using analog VLSI design tools, are interfaced to a PC. The building block chip designs, the cascadability and the hardware-in-the-loop supervised learning aspects of these chips are described.

Bio-inspired spiking neural network for nonlinear systems control.

PubMed

Pérez, Javier; Cabrera, Juan A; Castillo, Juan J; Velasco, Juan M

2018-08-01

Spiking neural networks (SNN) are the third generation of artificial neural networks. SNN are the closest approximation to biological neural networks. SNNs make use of temporal spike trains to command inputs and outputs, allowing a faster and more complex computation. As demonstrated by biological organisms, they are a potentially good approach to designing controllers for highly nonlinear dynamic systems in which the performance of controllers developed by conventional techniques is not satisfactory or difficult to implement. SNN-based controllers exploit their ability for online learning and self-adaptation to evolve when transferred from simulations to the real world. SNN's inherent binary and temporary way of information codification facilitates their hardware implementation compared to analog neurons. Biological neural networks often require a lower number of neurons compared to other controllers based on artificial neural networks. In this work, these neuronal systems are imitated to perform the control of non-linear dynamic systems. For this purpose, a control structure based on spiking neural networks has been designed. Particular attention has been paid to optimizing the structure and size of the neural network. The proposed structure is able to control dynamic systems with a reduced number of neurons and connections. A supervised learning process using evolutionary algorithms has been carried out to perform controller training. The efficiency of the proposed network has been verified in two examples of dynamic systems control. Simulations show that the proposed control based on SNN exhibits superior performance compared to other approaches based on Neural Networks and SNNs. Copyright © 2018 Elsevier Ltd. All rights reserved.

Deep Visual Attention Prediction

NASA Astrophysics Data System (ADS)

Wang, Wenguan; Shen, Jianbing

2018-05-01

In this work, we aim to predict human eye fixation with view-free scenes based on an end-to-end deep learning architecture. Although Convolutional Neural Networks (CNNs) have made substantial improvement on human attention prediction, it is still needed to improve CNN based attention models by efficiently leveraging multi-scale features. Our visual attention network is proposed to capture hierarchical saliency information from deep, coarse layers with global saliency information to shallow, fine layers with local saliency response. Our model is based on a skip-layer network structure, which predicts human attention from multiple convolutional layers with various reception fields. Final saliency prediction is achieved via the cooperation of those global and local predictions. Our model is learned in a deep supervision manner, where supervision is directly fed into multi-level layers, instead of previous approaches of providing supervision only at the output layer and propagating this supervision back to earlier layers. Our model thus incorporates multi-level saliency predictions within a single network, which significantly decreases the redundancy of previous approaches of learning multiple network streams with different input scales. Extensive experimental analysis on various challenging benchmark datasets demonstrate our method yields state-of-the-art performance with competitive inference time.

Neural-network classifiers for automatic real-world aerial image recognition

NASA Astrophysics Data System (ADS)

Greenberg, Shlomo; Guterman, Hugo

1996-08-01

We describe the application of the multilayer perceptron (MLP) network and a version of the adaptive resonance theory version 2-A (ART 2-A) network to the problem of automatic aerial image recognition (AAIR). The classification of aerial images, independent of their positions and orientations, is required for automatic tracking and target recognition. Invariance is achieved by the use of different invariant feature spaces in combination with supervised and unsupervised neural networks. The performance of neural-network-based classifiers in conjunction with several types of invariant AAIR global features, such as the Fourier-transform space, Zernike moments, central moments, and polar transforms, are examined. The advantages of this approach are discussed. The performance of the MLP network is compared with that of a classical correlator. The MLP neural-network correlator outperformed the binary phase-only filter (BPOF) correlator. It was found that the ART 2-A distinguished itself with its speed and its low number of required training vectors. However, only the MLP classifier was able to deal with a combination of shift and rotation geometric distortions.

Neural-network classifiers for automatic real-world aerial image recognition.

PubMed

Greenberg, S; Guterman, H

1996-08-10

We describe the application of the multilayer perceptron (MLP) network and a version of the adaptive resonance theory version 2-A (ART 2-A) network to the problem of automatic aerial image recognition (AAIR). The classification of aerial images, independent of their positions and orientations, is required for automatic tracking and target recognition. Invariance is achieved by the use of different invariant feature spaces in combination with supervised and unsupervised neural networks. The performance of neural-network-based classifiers in conjunction with several types of invariant AAIR global features, such as the Fourier-transform space, Zernike moments, central moments, and polar transforms, are examined. The advantages of this approach are discussed. The performance of the MLP network is compared with that of a classical correlator. The MLP neural-network correlator outperformed the binary phase-only filter (BPOF) correlator. It was found that the ART 2-A distinguished itself with its speed and its low number of required training vectors. However, only the MLP classifier was able to deal with a combination of shift and rotation geometric distortions.

Automated Detection of Clinically Significant Prostate Cancer in mp-MRI Images Based on an End-to-End Deep Neural Network.

PubMed

Wang, Zhiwei; Liu, Chaoyue; Cheng, Danpeng; Wang, Liang; Yang, Xin; Cheng, Kwang-Ting

2018-05-01

Automated methods for detecting clinically significant (CS) prostate cancer (PCa) in multi-parameter magnetic resonance images (mp-MRI) are of high demand. Existing methods typically employ several separate steps, each of which is optimized individually without considering the error tolerance of other steps. As a result, they could either involve unnecessary computational cost or suffer from errors accumulated over steps. In this paper, we present an automated CS PCa detection system, where all steps are optimized jointly in an end-to-end trainable deep neural network. The proposed neural network consists of concatenated subnets: 1) a novel tissue deformation network (TDN) for automated prostate detection and multimodal registration and 2) a dual-path convolutional neural network (CNN) for CS PCa detection. Three types of loss functions, i.e., classification loss, inconsistency loss, and overlap loss, are employed for optimizing all parameters of the proposed TDN and CNN. In the training phase, the two nets mutually affect each other and effectively guide registration and extraction of representative CS PCa-relevant features to achieve results with sufficient accuracy. The entire network is trained in a weakly supervised manner by providing only image-level annotations (i.e., presence/absence of PCa) without exact priors of lesions' locations. Compared with most existing systems which require supervised labels, e.g., manual delineation of PCa lesions, it is much more convenient for clinical usage. Comprehensive evaluation based on fivefold cross validation using 360 patient data demonstrates that our system achieves a high accuracy for CS PCa detection, i.e., a sensitivity of 0.6374 and 0.8978 at 0.1 and 1 false positives per normal/benign patient.

Predicting the survival of diabetes using neural network

NASA Astrophysics Data System (ADS)

Mamuda, Mamman; Sathasivam, Saratha

2017-08-01

Data mining techniques at the present time are used in predicting diseases of health care industries. Neural Network is one among the prevailing method in data mining techniques of an intelligent field for predicting diseases in health care industries. This paper presents a study on the prediction of the survival of diabetes diseases using different learning algorithms from the supervised learning algorithms of neural network. Three learning algorithms are considered in this study: (i) The levenberg-marquardt learning algorithm (ii) The Bayesian regulation learning algorithm and (iii) The scaled conjugate gradient learning algorithm. The network is trained using the Pima Indian Diabetes Dataset with the help of MATLAB R2014(a) software. The performance of each algorithm is further discussed through regression analysis. The prediction accuracy of the best algorithm is further computed to validate the accurate prediction

Rotation invariant deep binary hashing for fast image retrieval

NASA Astrophysics Data System (ADS)

Dai, Lai; Liu, Jianming; Jiang, Aiwen

2017-07-01

In this paper, we study how to compactly represent image's characteristics for fast image retrieval. We propose supervised rotation invariant compact discriminative binary descriptors through combining convolutional neural network with hashing. In the proposed network, binary codes are learned by employing a hidden layer for representing latent concepts that dominate on class labels. A loss function is proposed to minimize the difference between binary descriptors that describe reference image and the rotated one. Compared with some other supervised methods, the proposed network doesn't have to require pair-wised inputs for binary code learning. Experimental results show that our method is effective and achieves state-of-the-art results on the CIFAR-10 and MNIST datasets.

Use of a Neural Net to Model the Impact of Optical Coherence Tomography Abnormalities on Vision in Age-related Macular Degeneration.

PubMed

Aslam, Tariq M; Zaki, Haider R; Mahmood, Sajjad; Ali, Zaria C; Ahmad, Nur A; Thorell, Mariana R; Balaskas, Konstantinos

2018-01-01

To develop a neural network for the estimation of visual acuity from optical coherence tomography (OCT) images of patients with neovascular age-related macular degeneration (AMD) and to demonstrate its use to model the impact of specific controlled OCT changes on vision. Artificial intelligence (neural network) study. We assessed 1400 OCT scans of patients with neovascular AMD. Fifteen physical features for each eligible OCT, as well as patient age, were used as input data and corresponding recorded visual acuity as the target data to train, validate, and test a supervised neural network. We then applied this network to model the impact on acuity of defined OCT changes in subretinal fluid, subretinal hyperreflective material, and loss of external limiting membrane (ELM) integrity. A total of 1210 eligible OCT scans were analyzed, resulting in 1210 data points, which were each 16-dimensional. A 10-layer feed-forward neural network with 1 hidden layer of 10 neurons was trained to predict acuity and demonstrated a root mean square error of 8.2 letters for predicted compared to actual visual acuity and a mean regression coefficient of 0.85. A virtual model using this network demonstrated the relationship of visual acuity to specific, programmed changes in OCT characteristics. When ELM is intact, there is a shallow decline in acuity with increasing subretinal fluid but a much steeper decline with equivalent increasing subretinal hyperreflective material. When ELM is not intact, all visual acuities are reduced. Increasing subretinal hyperreflective material or subretinal fluid in this circumstance reduces vision further still, but with a smaller gradient than when ELM is intact. The supervised machine learning neural network developed is able to generate an estimated visual acuity value from OCT images in a population of patients with AMD. These findings should be of clinical and research interest in macular degeneration, for example in estimating visual prognosis or highlighting the importance of developing treatments targeting more visually destructive pathologies. Copyright © 2017 Elsevier Inc. All rights reserved.

A Deep Learning-Based Method for Similar Patient Question Retrieval in Chinese.

PubMed

Tang, Guo Yu; Ni, Yuan; Xie, Guo Tong; Fan, Xin Li; Shi, Yan Ling

2017-01-01

The online patient question and answering (Q&A) system, either as a website or a mobile application, attracts an increasing number of users in China. Patients will post their questions and the registered doctors then provide the corresponding answers. A large amount of questions with answers from doctors are accumulated. Instead of awaiting the response from a doctor, the newly posted question could be quickly answered by finding a semantically equivalent question from the Q&A achive. In this study, we investigated a novel deep learning based method to retrieve the similar patient question in Chinese. An unsupervised learning algorithm using deep neural network is performed on the corpus to generate the word embedding. The word embedding was then used as the input to a supervised learning algorithm using a designed deep neural network, i.e. the supervised neural attention model (SNA), to predict the similarity between two questions. The experimental results showed that our SNA method achieved P@1 = 77% and P@5 = 84%, which outperformed all other compared methods.

DCS-Neural-Network Program for Aircraft Control and Testing

NASA Technical Reports Server (NTRS)

Jorgensen, Charles C.

2006-01-01

A computer program implements a dynamic-cell-structure (DCS) artificial neural network that can perform such tasks as learning selected aerodynamic characteristics of an airplane from wind-tunnel test data and computing real-time stability and control derivatives of the airplane for use in feedback linearized control. A DCS neural network is one of several types of neural networks that can incorporate additional nodes in order to rapidly learn increasingly complex relationships between inputs and outputs. In the DCS neural network implemented by the present program, the insertion of nodes is based on accumulated error. A competitive Hebbian learning rule (a supervised-learning rule in which connection weights are adjusted to minimize differences between actual and desired outputs for training examples) is used. A Kohonen-style learning rule (derived from a relatively simple training algorithm, implements a Delaunay triangulation layout of neurons) is used to adjust node positions during training. Neighborhood topology determines which nodes are used to estimate new values. The network learns, starting with two nodes, and adds new nodes sequentially in locations chosen to maximize reductions in global error. At any given time during learning, the error becomes homogeneously distributed over all nodes.

Constrained Deep Weak Supervision for Histopathology Image Segmentation.

PubMed

Jia, Zhipeng; Huang, Xingyi; Chang, Eric I-Chao; Xu, Yan

2017-11-01

In this paper, we develop a new weakly supervised learning algorithm to learn to segment cancerous regions in histopathology images. This paper is under a multiple instance learning (MIL) framework with a new formulation, deep weak supervision (DWS); we also propose an effective way to introduce constraints to our neural networks to assist the learning process. The contributions of our algorithm are threefold: 1) we build an end-to-end learning system that segments cancerous regions with fully convolutional networks (FCNs) in which image-to-image weakly-supervised learning is performed; 2) we develop a DWS formulation to exploit multi-scale learning under weak supervision within FCNs; and 3) constraints about positive instances are introduced in our approach to effectively explore additional weakly supervised information that is easy to obtain and enjoy a significant boost to the learning process. The proposed algorithm, abbreviated as DWS-MIL, is easy to implement and can be trained efficiently. Our system demonstrates the state-of-the-art results on large-scale histopathology image data sets and can be applied to various applications in medical imaging beyond histopathology images, such as MRI, CT, and ultrasound images.

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.

Mastering the game of Go without human knowledge.

PubMed

Silver, David; Schrittwieser, Julian; Simonyan, Karen; Antonoglou, Ioannis; Huang, Aja; Guez, Arthur; Hubert, Thomas; Baker, Lucas; Lai, Matthew; Bolton, Adrian; Chen, Yutian; Lillicrap, Timothy; Hui, Fan; Sifre, Laurent; van den Driessche, George; Graepel, Thore; Hassabis, Demis

2017-10-18

A long-standing goal of artificial intelligence is an algorithm that learns, tabula rasa, superhuman proficiency in challenging domains. Recently, AlphaGo became the first program to defeat a world champion in the game of Go. The tree search in AlphaGo evaluated positions and selected moves using deep neural networks. These neural networks were trained by supervised learning from human expert moves, and by reinforcement learning from self-play. Here we introduce an algorithm based solely on reinforcement learning, without human data, guidance or domain knowledge beyond game rules. AlphaGo becomes its own teacher: a neural network is trained to predict AlphaGo's own move selections and also the winner of AlphaGo's games. This neural network improves the strength of the tree search, resulting in higher quality move selection and stronger self-play in the next iteration. Starting tabula rasa, our new program AlphaGo Zero achieved superhuman performance, winning 100-0 against the previously published, champion-defeating AlphaGo.

Mastering the game of Go without human knowledge

NASA Astrophysics Data System (ADS)

Silver, David; Schrittwieser, Julian; Simonyan, Karen; Antonoglou, Ioannis; Huang, Aja; Guez, Arthur; Hubert, Thomas; Baker, Lucas; Lai, Matthew; Bolton, Adrian; Chen, Yutian; Lillicrap, Timothy; Hui, Fan; Sifre, Laurent; van den Driessche, George; Graepel, Thore; Hassabis, Demis

2017-10-01

A long-standing goal of artificial intelligence is an algorithm that learns, tabula rasa, superhuman proficiency in challenging domains. Recently, AlphaGo became the first program to defeat a world champion in the game of Go. The tree search in AlphaGo evaluated positions and selected moves using deep neural networks. These neural networks were trained by supervised learning from human expert moves, and by reinforcement learning from self-play. Here we introduce an algorithm based solely on reinforcement learning, without human data, guidance or domain knowledge beyond game rules. AlphaGo becomes its own teacher: a neural network is trained to predict AlphaGo’s own move selections and also the winner of AlphaGo’s games. This neural network improves the strength of the tree search, resulting in higher quality move selection and stronger self-play in the next iteration. Starting tabula rasa, our new program AlphaGo Zero achieved superhuman performance, winning 100-0 against the previously published, champion-defeating AlphaGo.

Evaluation of supervised machine-learning algorithms to distinguish between inflammatory bowel disease and alimentary lymphoma in cats.

PubMed

Awaysheh, Abdullah; Wilcke, Jeffrey; Elvinger, François; Rees, Loren; Fan, Weiguo; Zimmerman, Kurt L

2016-11-01

Inflammatory bowel disease (IBD) and alimentary lymphoma (ALA) are common gastrointestinal diseases in cats. The very similar clinical signs and histopathologic features of these diseases make the distinction between them diagnostically challenging. We tested the use of supervised machine-learning algorithms to differentiate between the 2 diseases using data generated from noninvasive diagnostic tests. Three prediction models were developed using 3 machine-learning algorithms: naive Bayes, decision trees, and artificial neural networks. The models were trained and tested on data from complete blood count (CBC) and serum chemistry (SC) results for the following 3 groups of client-owned cats: normal, inflammatory bowel disease (IBD), or alimentary lymphoma (ALA). Naive Bayes and artificial neural networks achieved higher classification accuracy (sensitivities of 70.8% and 69.2%, respectively) than the decision tree algorithm (63%, p < 0.0001). The areas under the receiver-operating characteristic curve for classifying cases into the 3 categories was 83% by naive Bayes, 79% by decision tree, and 82% by artificial neural networks. Prediction models using machine learning provided a method for distinguishing between ALA-IBD, ALA-normal, and IBD-normal. The naive Bayes and artificial neural networks classifiers used 10 and 4 of the CBC and SC variables, respectively, to outperform the C4.5 decision tree, which used 5 CBC and SC variables in classifying cats into the 3 classes. These models can provide another noninvasive diagnostic tool to assist clinicians with differentiating between IBD and ALA, and between diseased and nondiseased cats. © 2016 The Author(s).

Deep Unfolding for Topic Models.

PubMed

Chien, Jen-Tzung; Lee, Chao-Hsi

2018-02-01

Deep unfolding provides an approach to integrate the probabilistic generative models and the deterministic neural networks. Such an approach is benefited by deep representation, easy interpretation, flexible learning and stochastic modeling. This study develops the unsupervised and supervised learning of deep unfolded topic models for document representation and classification. Conventionally, the unsupervised and supervised topic models are inferred via the variational inference algorithm where the model parameters are estimated by maximizing the lower bound of logarithm of marginal likelihood using input documents without and with class labels, respectively. The representation capability or classification accuracy is constrained by the variational lower bound and the tied model parameters across inference procedure. This paper aims to relax these constraints by directly maximizing the end performance criterion and continuously untying the parameters in learning process via deep unfolding inference (DUI). The inference procedure is treated as the layer-wise learning in a deep neural network. The end performance is iteratively improved by using the estimated topic parameters according to the exponentiated updates. Deep learning of topic models is therefore implemented through a back-propagation procedure. Experimental results show the merits of DUI with increasing number of layers compared with variational inference in unsupervised as well as supervised topic models.

Deep learning for brain tumor classification

NASA Astrophysics Data System (ADS)

Paul, Justin S.; Plassard, Andrew J.; Landman, Bennett A.; Fabbri, Daniel

2017-03-01

Recent research has shown that deep learning methods have performed well on supervised machine learning, image classification tasks. The purpose of this study is to apply deep learning methods to classify brain images with different tumor types: meningioma, glioma, and pituitary. A dataset was publicly released containing 3,064 T1-weighted contrast enhanced MRI (CE-MRI) brain images from 233 patients with either meningioma, glioma, or pituitary tumors split across axial, coronal, or sagittal planes. This research focuses on the 989 axial images from 191 patients in order to avoid confusing the neural networks with three different planes containing the same diagnosis. Two types of neural networks were used in classification: fully connected and convolutional neural networks. Within these two categories, further tests were computed via the augmentation of the original 512×512 axial images. Training neural networks over the axial data has proven to be accurate in its classifications with an average five-fold cross validation of 91.43% on the best trained neural network. This result demonstrates that a more general method (i.e. deep learning) can outperform specialized methods that require image dilation and ring-forming subregions on tumors.

Segmentation of retinal blood vessels using artificial neural networks for early detection of diabetic retinopathy

NASA Astrophysics Data System (ADS)

Mann, Kulwinder S.; Kaur, Sukhpreet

2017-06-01

There are various eye diseases in the patients suffering from the diabetes which includes Diabetic Retinopathy, Glaucoma, Hypertension etc. These all are the most common sight threatening eye diseases due to the changes in the blood vessel structure. The proposed method using supervised methods concluded that the segmentation of the retinal blood vessels can be performed accurately using neural networks training. It uses features which include Gray level features; Moment Invariant based features, Gabor filtering, Intensity feature, Vesselness feature for feature vector computation. Then the feature vector is calculated using only the prominent features.

Self-supervised ARTMAP.

PubMed

Amis, Gregory P; Carpenter, Gail A

2010-03-01

Computational models of learning typically train on labeled input patterns (supervised learning), unlabeled input patterns (unsupervised learning), or a combination of the two (semi-supervised learning). In each case input patterns have a fixed number of features throughout training and testing. Human and machine learning contexts present additional opportunities for expanding incomplete knowledge from formal training, via self-directed learning that incorporates features not previously experienced. This article defines a new self-supervised learning paradigm to address these richer learning contexts, introducing a neural network called self-supervised ARTMAP. Self-supervised learning integrates knowledge from a teacher (labeled patterns with some features), knowledge from the environment (unlabeled patterns with more features), and knowledge from internal model activation (self-labeled patterns). Self-supervised ARTMAP learns about novel features from unlabeled patterns without destroying partial knowledge previously acquired from labeled patterns. A category selection function bases system predictions on known features, and distributed network activation scales unlabeled learning to prediction confidence. Slow distributed learning on unlabeled patterns focuses on novel features and confident predictions, defining classification boundaries that were ambiguous in the labeled patterns. Self-supervised ARTMAP improves test accuracy on illustrative low-dimensional problems and on high-dimensional benchmarks. Model code and benchmark data are available from: http://techlab.eu.edu/SSART/. Copyright 2009 Elsevier Ltd. All rights reserved.

Convergence analysis of sliding mode trajectories in multi-objective neural networks learning.

PubMed

Costa, Marcelo Azevedo; Braga, Antonio Padua; de Menezes, Benjamin Rodrigues

2012-09-01

The Pareto-optimality concept is used in this paper in order to represent a constrained set of solutions that are able to trade-off the two main objective functions involved in neural networks supervised learning: data-set error and network complexity. The neural network is described as a dynamic system having error and complexity as its state variables and learning is presented as a process of controlling a learning trajectory in the resulting state space. In order to control the trajectories, sliding mode dynamics is imposed to the network. It is shown that arbitrary learning trajectories can be achieved by maintaining the sliding mode gains within their convergence intervals. Formal proofs of convergence conditions are therefore presented. The concept of trajectory learning presented in this paper goes further beyond the selection of a final state in the Pareto set, since it can be reached through different trajectories and states in the trajectory can be assessed individually against an additional objective function. Copyright © 2012 Elsevier Ltd. All rights reserved.

Semantic Segmentation of Convolutional Neural Network for Supervised Classification of Multispectral Remote Sensing

NASA Astrophysics Data System (ADS)

Xue, L.; Liu, C.; Wu, Y.; Li, H.

2018-04-01

Semantic segmentation is a fundamental research in remote sensing image processing. Because of the complex maritime environment, the classification of roads, vegetation, buildings and water from remote Sensing Imagery is a challenging task. Although the neural network has achieved excellent performance in semantic segmentation in the last years, there are a few of works using CNN for ground object segmentation and the results could be further improved. This paper used convolution neural network named U-Net, its structure has a contracting path and an expansive path to get high resolution output. In the network , We added BN layers, which is more conducive to the reverse pass. Moreover, after upsampling convolution , we add dropout layers to prevent overfitting. They are promoted to get more precise segmentation results. To verify this network architecture, we used a Kaggle dataset. Experimental results show that U-Net achieved good performance compared with other architectures, especially in high-resolution remote sensing imagery.

Hybrid computing using a neural network with dynamic external memory.

PubMed

Graves, Alex; Wayne, Greg; Reynolds, Malcolm; Harley, Tim; Danihelka, Ivo; Grabska-Barwińska, Agnieszka; Colmenarejo, Sergio Gómez; Grefenstette, Edward; Ramalho, Tiago; Agapiou, John; Badia, Adrià Puigdomènech; Hermann, Karl Moritz; Zwols, Yori; Ostrovski, Georg; Cain, Adam; King, Helen; Summerfield, Christopher; Blunsom, Phil; Kavukcuoglu, Koray; Hassabis, Demis

2016-10-27

Artificial neural networks are remarkably adept at sensory processing, sequence learning and reinforcement learning, but are limited in their ability to represent variables and data structures and to store data over long timescales, owing to the lack of an external memory. Here we introduce a machine learning model called a differentiable neural computer (DNC), which consists of a neural network that can read from and write to an external memory matrix, analogous to the random-access memory in a conventional computer. Like a conventional computer, it can use its memory to represent and manipulate complex data structures, but, like a neural network, it can learn to do so from data. When trained with supervised learning, we demonstrate that a DNC can successfully answer synthetic questions designed to emulate reasoning and inference problems in natural language. We show that it can learn tasks such as finding the shortest path between specified points and inferring the missing links in randomly generated graphs, and then generalize these tasks to specific graphs such as transport networks and family trees. When trained with reinforcement learning, a DNC can complete a moving blocks puzzle in which changing goals are specified by sequences of symbols. Taken together, our results demonstrate that DNCs have the capacity to solve complex, structured tasks that are inaccessible to neural networks without external read-write memory.

A link prediction method for heterogeneous networks based on BP neural network

NASA Astrophysics Data System (ADS)

Li, Ji-chao; Zhao, Dan-ling; Ge, Bing-Feng; Yang, Ke-Wei; Chen, Ying-Wu

2018-04-01

Most real-world systems, composed of different types of objects connected via many interconnections, can be abstracted as various complex heterogeneous networks. Link prediction for heterogeneous networks is of great significance for mining missing links and reconfiguring networks according to observed information, with considerable applications in, for example, friend and location recommendations and disease-gene candidate detection. In this paper, we put forward a novel integrated framework, called MPBP (Meta-Path feature-based BP neural network model), to predict multiple types of links for heterogeneous networks. More specifically, the concept of meta-path is introduced, followed by the extraction of meta-path features for heterogeneous networks. Next, based on the extracted meta-path features, a supervised link prediction model is built with a three-layer BP neural network. Then, the solution algorithm of the proposed link prediction model is put forward to obtain predicted results by iteratively training the network. Last, numerical experiments on the dataset of examples of a gene-disease network and a combat network are conducted to verify the effectiveness and feasibility of the proposed MPBP. It shows that the MPBP with very good performance is superior to the baseline methods.

Neural computing for numeric-to-symbolic conversion in control systems

NASA Technical Reports Server (NTRS)

Passino, Kevin M.; Sartori, Michael A.; Antsaklis, Panos J.

1989-01-01

A type of neural network, the multilayer perceptron, is used to classify numeric data and assign appropriate symbols to various classes. This numeric-to-symbolic conversion results in a type of information extraction, which is similar to what is called data reduction in pattern recognition. The use of the neural network as a numeric-to-symbolic converter is introduced, its application in autonomous control is discussed, and several applications are studied. The perceptron is used as a numeric-to-symbolic converter for a discrete-event system controller supervising a continuous variable dynamic system. It is also shown how the perceptron can implement fault trees, which provide useful information (alarms) in a biological system and information for failure diagnosis and control purposes in an aircraft example.

Event Recognition Based on Deep Learning in Chinese Texts

PubMed Central

Zhang, Yajun; Liu, Zongtian; Zhou, Wen

2016-01-01

Event recognition is the most fundamental and critical task in event-based natural language processing systems. Existing event recognition methods based on rules and shallow neural networks have certain limitations. For example, extracting features using methods based on rules is difficult; methods based on shallow neural networks converge too quickly to a local minimum, resulting in low recognition precision. To address these problems, we propose the Chinese emergency event recognition model based on deep learning (CEERM). Firstly, we use a word segmentation system to segment sentences. According to event elements labeled in the CEC 2.0 corpus, we classify words into five categories: trigger words, participants, objects, time and location. Each word is vectorized according to the following six feature layers: part of speech, dependency grammar, length, location, distance between trigger word and core word and trigger word frequency. We obtain deep semantic features of words by training a feature vector set using a deep belief network (DBN), then analyze those features in order to identify trigger words by means of a back propagation neural network. Extensive testing shows that the CEERM achieves excellent recognition performance, with a maximum F-measure value of 85.17%. Moreover, we propose the dynamic-supervised DBN, which adds supervised fine-tuning to a restricted Boltzmann machine layer by monitoring its training performance. Test analysis reveals that the new DBN improves recognition performance and effectively controls the training time. Although the F-measure increases to 88.11%, the training time increases by only 25.35%. PMID:27501231

Microcrack Quantification in Composite Materials by a Neural Network Analysis of Ultrasound Spectral Data

NASA Technical Reports Server (NTRS)

Walker, James L.; Russell, Samuel S.; Suits, Michael W.

2003-01-01

Intra-ply microcracking in unlined composite pressure vessels can be very troublesome to detect and when linked through the thickness can provide leak paths that may hinder mission success. The leaks may lead to loss of pressure/propellant, increased risk of explosion and possible cryo-pumping into air pockets within the laminate. Ultrasonic techniques have been shown capable of detecting the presence of microcracking and in this work they are used to quantify the level of microcracking. Resonance ultrasound methods are utilized with artificial neural networks to build a microcrack prediction/measurement tool. Two networks are presented, one unsupervised to provide a qualitative measure of microcracking and one supervised which provides a quantitative assessment of the level of microcracking. The resonant ultrasound spectroscopic method is made sensitive to microcracking by tuning the input spectrum to the higher frequency (shorter wavelength) components allowing more significant interaction with the defects. This interaction causes the spectral characteristics to shift toward lower amplitudes at the higher frequencies. As the density of the defects increases more interactions occur and more drastic amplitude changes are observed. Preliminary experiments to quantify the level of microcracking induced in graphite/epoxy composite samples through a combination of tensile loading and cryogenic temperatures are presented. Both unsupervised (Kohonen) and supervised (radial basis function) artificial neural networks are presented to determine the measurable effect on the resonance spectrum of the ultrasonic data taken from the samples.

Event Recognition Based on Deep Learning in Chinese Texts.

PubMed

Zhang, Yajun; Liu, Zongtian; Zhou, Wen

2016-01-01

Event recognition is the most fundamental and critical task in event-based natural language processing systems. Existing event recognition methods based on rules and shallow neural networks have certain limitations. For example, extracting features using methods based on rules is difficult; methods based on shallow neural networks converge too quickly to a local minimum, resulting in low recognition precision. To address these problems, we propose the Chinese emergency event recognition model based on deep learning (CEERM). Firstly, we use a word segmentation system to segment sentences. According to event elements labeled in the CEC 2.0 corpus, we classify words into five categories: trigger words, participants, objects, time and location. Each word is vectorized according to the following six feature layers: part of speech, dependency grammar, length, location, distance between trigger word and core word and trigger word frequency. We obtain deep semantic features of words by training a feature vector set using a deep belief network (DBN), then analyze those features in order to identify trigger words by means of a back propagation neural network. Extensive testing shows that the CEERM achieves excellent recognition performance, with a maximum F-measure value of 85.17%. Moreover, we propose the dynamic-supervised DBN, which adds supervised fine-tuning to a restricted Boltzmann machine layer by monitoring its training performance. Test analysis reveals that the new DBN improves recognition performance and effectively controls the training time. Although the F-measure increases to 88.11%, the training time increases by only 25.35%.

Learning the ideal observer for SKE detection tasks by use of convolutional neural networks (Cum Laude Poster Award)

NASA Astrophysics Data System (ADS)

Zhou, Weimin; Anastasio, Mark A.

2018-03-01

It has been advocated that task-based measures of image quality (IQ) should be employed to evaluate and optimize imaging systems. Task-based measures of IQ quantify the performance of an observer on a medically relevant task. The Bayesian Ideal Observer (IO), which employs complete statistical information of the object and noise, achieves the upper limit of the performance for a binary signal classification task. However, computing the IO performance is generally analytically intractable and can be computationally burdensome when Markov-chain Monte Carlo (MCMC) techniques are employed. In this paper, supervised learning with convolutional neural networks (CNNs) is employed to approximate the IO test statistics for a signal-known-exactly and background-known-exactly (SKE/BKE) binary detection task. The receiver operating characteristic (ROC) curve and the area under the ROC curve (AUC) are compared to those produced by the analytically computed IO. The advantages of the proposed supervised learning approach for approximating the IO are demonstrated.

An Event-Driven Classifier for Spiking Neural Networks Fed with Synthetic or Dynamic Vision Sensor Data.

PubMed

Stromatias, Evangelos; Soto, Miguel; Serrano-Gotarredona, Teresa; Linares-Barranco, Bernabé

2017-01-01

This paper introduces a novel methodology for training an event-driven classifier within a Spiking Neural Network (SNN) System capable of yielding good classification results when using both synthetic input data and real data captured from Dynamic Vision Sensor (DVS) chips. The proposed supervised method uses the spiking activity provided by an arbitrary topology of prior SNN layers to build histograms and train the classifier in the frame domain using the stochastic gradient descent algorithm. In addition, this approach can cope with leaky integrate-and-fire neuron models within the SNN, a desirable feature for real-world SNN applications, where neural activation must fade away after some time in the absence of inputs. Consequently, this way of building histograms captures the dynamics of spikes immediately before the classifier. We tested our method on the MNIST data set using different synthetic encodings and real DVS sensory data sets such as N-MNIST, MNIST-DVS, and Poker-DVS using the same network topology and feature maps. We demonstrate the effectiveness of our approach by achieving the highest classification accuracy reported on the N-MNIST (97.77%) and Poker-DVS (100%) real DVS data sets to date with a spiking convolutional network. Moreover, by using the proposed method we were able to retrain the output layer of a previously reported spiking neural network and increase its performance by 2%, suggesting that the proposed classifier can be used as the output layer in works where features are extracted using unsupervised spike-based learning methods. In addition, we also analyze SNN performance figures such as total event activity and network latencies, which are relevant for eventual hardware implementations. In summary, the paper aggregates unsupervised-trained SNNs with a supervised-trained SNN classifier, combining and applying them to heterogeneous sets of benchmarks, both synthetic and from real DVS chips.

Supervised Learning in Spiking Neural Networks for Precise Temporal Encoding

PubMed Central

Gardner, Brian; Grüning, André

2016-01-01

Precise spike timing as a means to encode information in neural networks is biologically supported, and is advantageous over frequency-based codes by processing input features on a much shorter time-scale. For these reasons, much recent attention has been focused on the development of supervised learning rules for spiking neural networks that utilise a temporal coding scheme. However, despite significant progress in this area, there still lack rules that have a theoretical basis, and yet can be considered biologically relevant. Here we examine the general conditions under which synaptic plasticity most effectively takes place to support the supervised learning of a precise temporal code. As part of our analysis we examine two spike-based learning methods: one of which relies on an instantaneous error signal to modify synaptic weights in a network (INST rule), and the other one relying on a filtered error signal for smoother synaptic weight modifications (FILT rule). We test the accuracy of the solutions provided by each rule with respect to their temporal encoding precision, and then measure the maximum number of input patterns they can learn to memorise using the precise timings of individual spikes as an indication of their storage capacity. Our results demonstrate the high performance of the FILT rule in most cases, underpinned by the rule’s error-filtering mechanism, which is predicted to provide smooth convergence towards a desired solution during learning. We also find the FILT rule to be most efficient at performing input pattern memorisations, and most noticeably when patterns are identified using spikes with sub-millisecond temporal precision. In comparison with existing work, we determine the performance of the FILT rule to be consistent with that of the highly efficient E-learning Chronotron rule, but with the distinct advantage that our FILT rule is also implementable as an online method for increased biological realism. PMID:27532262

Supervised Learning in Spiking Neural Networks for Precise Temporal Encoding.

PubMed

Gardner, Brian; Grüning, André

2016-01-01

Precise spike timing as a means to encode information in neural networks is biologically supported, and is advantageous over frequency-based codes by processing input features on a much shorter time-scale. For these reasons, much recent attention has been focused on the development of supervised learning rules for spiking neural networks that utilise a temporal coding scheme. However, despite significant progress in this area, there still lack rules that have a theoretical basis, and yet can be considered biologically relevant. Here we examine the general conditions under which synaptic plasticity most effectively takes place to support the supervised learning of a precise temporal code. As part of our analysis we examine two spike-based learning methods: one of which relies on an instantaneous error signal to modify synaptic weights in a network (INST rule), and the other one relying on a filtered error signal for smoother synaptic weight modifications (FILT rule). We test the accuracy of the solutions provided by each rule with respect to their temporal encoding precision, and then measure the maximum number of input patterns they can learn to memorise using the precise timings of individual spikes as an indication of their storage capacity. Our results demonstrate the high performance of the FILT rule in most cases, underpinned by the rule's error-filtering mechanism, which is predicted to provide smooth convergence towards a desired solution during learning. We also find the FILT rule to be most efficient at performing input pattern memorisations, and most noticeably when patterns are identified using spikes with sub-millisecond temporal precision. In comparison with existing work, we determine the performance of the FILT rule to be consistent with that of the highly efficient E-learning Chronotron rule, but with the distinct advantage that our FILT rule is also implementable as an online method for increased biological realism.

End-to-End ASR-Free Keyword Search From Speech

NASA Astrophysics Data System (ADS)

Audhkhasi, Kartik; Rosenberg, Andrew; Sethy, Abhinav; Ramabhadran, Bhuvana; Kingsbury, Brian

2017-12-01

End-to-end (E2E) systems have achieved competitive results compared to conventional hybrid hidden Markov model (HMM)-deep neural network based automatic speech recognition (ASR) systems. Such E2E systems are attractive due to the lack of dependence on alignments between input acoustic and output grapheme or HMM state sequence during training. This paper explores the design of an ASR-free end-to-end system for text query-based keyword search (KWS) from speech trained with minimal supervision. Our E2E KWS system consists of three sub-systems. The first sub-system is a recurrent neural network (RNN)-based acoustic auto-encoder trained to reconstruct the audio through a finite-dimensional representation. The second sub-system is a character-level RNN language model using embeddings learned from a convolutional neural network. Since the acoustic and text query embeddings occupy different representation spaces, they are input to a third feed-forward neural network that predicts whether the query occurs in the acoustic utterance or not. This E2E ASR-free KWS system performs respectably despite lacking a conventional ASR system and trains much faster.

Training strategy for convolutional neural networks in pedestrian gender classification

NASA Astrophysics Data System (ADS)

Ng, Choon-Boon; Tay, Yong-Haur; Goi, Bok-Min

2017-06-01

In this work, we studied a strategy for training a convolutional neural network in pedestrian gender classification with limited amount of labeled training data. Unsupervised learning by k-means clustering on pedestrian images was used to learn the filters to initialize the first layer of the network. As a form of pre-training, supervised learning for the related task of pedestrian classification was performed. Finally, the network was fine-tuned for gender classification. We found that this strategy improved the network's generalization ability in gender classification, achieving better test results when compared to random weights initialization and slightly more beneficial than merely initializing the first layer filters by unsupervised learning. This shows that unsupervised learning followed by pre-training with pedestrian images is an effective strategy to learn useful features for pedestrian gender classification.

Multi-objective evolutionary optimization for constructing neural networks for virtual reality visual data mining: application to geophysical prospecting.

PubMed

Valdés, Julio J; Barton, Alan J

2007-05-01

A method for the construction of virtual reality spaces for visual data mining using multi-objective optimization with genetic algorithms on nonlinear discriminant (NDA) neural networks is presented. Two neural network layers (the output and the last hidden) are used for the construction of simultaneous solutions for: (i) a supervised classification of data patterns and (ii) an unsupervised similarity structure preservation between the original data matrix and its image in the new space. A set of spaces are constructed from selected solutions along the Pareto front. This strategy represents a conceptual improvement over spaces computed by single-objective optimization. In addition, genetic programming (in particular gene expression programming) is used for finding analytic representations of the complex mappings generating the spaces (a composition of NDA and orthogonal principal components). The presented approach is domain independent and is illustrated via application to the geophysical prospecting of caves.

Relabeling exchange method (REM) for learning in neural networks

NASA Astrophysics Data System (ADS)

Wu, Wen; Mammone, Richard J.

1994-02-01

The supervised training of neural networks require the use of output labels which are usually arbitrarily assigned. In this paper it is shown that there is a significant difference in the rms error of learning when `optimal' label assignment schemes are used. We have investigated two efficient random search algorithms to solve the relabeling problem: the simulated annealing and the genetic algorithm. However, we found them to be computationally expensive. Therefore we shall introduce a new heuristic algorithm called the Relabeling Exchange Method (REM) which is computationally more attractive and produces optimal performance. REM has been used to organize the optimal structure for multi-layered perceptrons and neural tree networks. The method is a general one and can be implemented as a modification to standard training algorithms. The motivation of the new relabeling strategy is based on the present interpretation of dyslexia as an encoding problem.

A neural network approach for image reconstruction in electron magnetic resonance tomography.

PubMed

Durairaj, D Christopher; Krishna, Murali C; Murugesan, Ramachandran

2007-10-01

An object-oriented, artificial neural network (ANN) based, application system for reconstruction of two-dimensional spatial images in electron magnetic resonance (EMR) tomography is presented. The standard back propagation algorithm is utilized to train a three-layer sigmoidal feed-forward, supervised, ANN to perform the image reconstruction. The network learns the relationship between the 'ideal' images that are reconstructed using filtered back projection (FBP) technique and the corresponding projection data (sinograms). The input layer of the network is provided with a training set that contains projection data from various phantoms as well as in vivo objects, acquired from an EMR imager. Twenty five different network configurations are investigated to test the ability of the generalization of the network. The trained ANN then reconstructs two-dimensional temporal spatial images that present the distribution of free radicals in biological systems. Image reconstruction by the trained neural network shows better time complexity than the conventional iterative reconstruction algorithms such as multiplicative algebraic reconstruction technique (MART). The network is further explored for image reconstruction from 'noisy' EMR data and the results show better performance than the FBP method. The network is also tested for its ability to reconstruct from limited-angle EMR data set.

Self-organizing neural networks--an alternative way of cluster analysis in clinical chemistry.

PubMed

Reibnegger, G; Wachter, H

1996-04-15

Supervised learning schemes have been employed by several workers for training neural networks designed to solve clinical problems. We demonstrate that unsupervised techniques can also produce interesting and meaningful results. Using a data set on the chemical composition of milk from 22 different mammals, we demonstrate that self-organizing feature maps (Kohonen networks) as well as a modified version of error backpropagation technique yield results mimicking conventional cluster analysis. Both techniques are able to project a potentially multi-dimensional input vector onto a two-dimensional space whereby neighborhood relationships remain conserved. Thus, these techniques can be used for reducing dimensionality of complicated data sets and for enhancing comprehensibility of features hidden in the data matrix.

Deep learning classification in asteroseismology

NASA Astrophysics Data System (ADS)

Hon, Marc; Stello, Dennis; Yu, Jie

2017-08-01

In the power spectra of oscillating red giants, there are visually distinct features defining stars ascending the red giant branch from those that have commenced helium core burning. We train a 1D convolutional neural network by supervised learning to automatically learn these visual features from images of folded oscillation spectra. By training and testing on Kepler red giants, we achieve an accuracy of up to 99 per cent in separating helium-burning red giants from those ascending the red giant branch. The convolutional neural network additionally shows capability in accurately predicting the evolutionary states of 5379 previously unclassified Kepler red giants, by which we now have greatly increased the number of classified stars.

Digital implementation of a neural network for imaging

NASA Astrophysics Data System (ADS)

Wood, Richard; McGlashan, Alex; Yatulis, Jay; Mascher, Peter; Bruce, Ian

2012-10-01

This paper outlines the design and testing of a digital imaging system that utilizes an artificial neural network with unsupervised and supervised learning to convert streaming input (real time) image space into parameter space. The primary objective of this work is to investigate the effectiveness of using a neural network to significantly reduce the information density of streaming images so that objects can be readily identified by a limited set of primary parameters and act as an enhanced human machine interface (HMI). Many applications are envisioned including use in biomedical imaging, anomaly detection and as an assistive device for the visually impaired. A digital circuit was designed and tested using a Field Programmable Gate Array (FPGA) and an off the shelf digital camera. Our results indicate that the networks can be readily trained when subject to limited sets of objects such as the alphabet. We can also separate limited object sets with rotational and positional invariance. The results also show that limited visual fields form with only local connectivity.

Supervised local error estimation for nonlinear image registration using convolutional neural networks

NASA Astrophysics Data System (ADS)

Eppenhof, Koen A. J.; Pluim, Josien P. W.

2017-02-01

Error estimation in medical image registration is valuable when validating, comparing, or combining registration methods. To validate a nonlinear image registration method, ideally the registration error should be known for the entire image domain. We propose a supervised method for the estimation of a registration error map for nonlinear image registration. The method is based on a convolutional neural network that estimates the norm of the residual deformation from patches around each pixel in two registered images. This norm is interpreted as the registration error, and is defined for every pixel in the image domain. The network is trained using a set of artificially deformed images. Each training example is a pair of images: the original image, and a random deformation of that image. No manually labeled ground truth error is required. At test time, only the two registered images are required as input. We train and validate the network on registrations in a set of 2D digital subtraction angiography sequences, such that errors up to eight pixels can be estimated. We show that for this range of errors the convolutional network is able to learn the registration error in pairs of 2D registered images at subpixel precision. Finally, we present a proof of principle for the extension to 3D registration problems in chest CTs, showing that the method has the potential to estimate errors in 3D registration problems.

Vibration control of building structures using self-organizing and self-learning neural networks

NASA Astrophysics Data System (ADS)

Madan, Alok

2005-11-01

Past research in artificial intelligence establishes that artificial neural networks (ANN) are effective and efficient computational processors for performing a variety of tasks including pattern recognition, classification, associative recall, combinatorial problem solving, adaptive control, multi-sensor data fusion, noise filtering and data compression, modelling and forecasting. The paper presents a potentially feasible approach for training ANN in active control of earthquake-induced vibrations in building structures without the aid of teacher signals (i.e. target control forces). A counter-propagation neural network is trained to output the control forces that are required to reduce the structural vibrations in the absence of any feedback on the correctness of the output control forces (i.e. without any information on the errors in output activations of the network). The present study shows that, in principle, the counter-propagation network (CPN) can learn from the control environment to compute the required control forces without the supervision of a teacher (unsupervised learning). Simulated case studies are presented to demonstrate the feasibility of implementing the unsupervised learning approach in ANN for effective vibration control of structures under the influence of earthquake ground motions. The proposed learning methodology obviates the need for developing a mathematical model of structural dynamics or training a separate neural network to emulate the structural response for implementation in practice.

Classification of asteroid spectra using a neural network

NASA Technical Reports Server (NTRS)

Howell, E. S.; Merenyi, E.; Lebofsky, L. A.

1994-01-01

The 52-color asteroid survey (Bell et al., 1988) together with the 8-color asteroid survey (Zellner et al., 1985) provide a data set of asteroid spectra spanning 0.3-2.5 micrometers. An artificial neural network clusters these asteroid spectra based on their similarity to each other. We have also trained the neural network with a categorization learning output layer in a supervised mode to associate the established clusters with taxonomic classes. Results of our classification agree with Tholen's classification based on the 8-color data alone. When extending the spectral range using the 52-color survey data, we find that some modification of the Tholen classes is indicated to produce a cleaner, self-consistent set of taxonomic classes. After supervised training using our modified classes, the network correctly classifies both the training examples, and additional spectra into the correct class with an average of 90% accuracy. Our classification supports the separation of the K class from the S class, as suggested by Bell et al. (1987), based on the near-infrared spectrum. We define two end-member subclasses which seem to have compositional significance within the S class: the So class, which is olivine-rich and red, and the Sp class, which is pyroxene-rich and less red. The remaining S-class asteroids have intermediate compositions of both olivine and pyroxene and moderately red continua. The network clustering suggests some additional structure within the E-, M-, and P-class asteroids, even in the absence of albedo information, which is the only discriminant between these in the Tholen classification. New relationships are seen between the C class and related G, B, and F classes. However, in both cases, the number of spectra is too small to interpret or determine the significance of these separations.

Teaching artificial neural systems to drive: Manual training techniques for autonomous systems

NASA Technical Reports Server (NTRS)

Shepanski, J. F.; Macy, S. A.

1987-01-01

A methodology was developed for manually training autonomous control systems based on artificial neural systems (ANS). In applications where the rule set governing an expert's decisions is difficult to formulate, ANS can be used to extract rules by associating the information an expert receives with the actions taken. Properly constructed networks imitate rules of behavior that permits them to function autonomously when they are trained on the spanning set of possible situations. This training can be provided manually, either under the direct supervision of a system trainer, or indirectly using a background mode where the networks assimilates training data as the expert performs its day-to-day tasks. To demonstrate these methods, an ANS network was trained to drive a vehicle through simulated freeway traffic.

A neural network approach for the blind deconvolution of turbulent flows

NASA Astrophysics Data System (ADS)

Maulik, R.; San, O.

2017-11-01

We present a single-layer feedforward artificial neural network architecture trained through a supervised learning approach for the deconvolution of flow variables from their coarse grained computations such as those encountered in large eddy simulations. We stress that the deconvolution procedure proposed in this investigation is blind, i.e. the deconvolved field is computed without any pre-existing information about the filtering procedure or kernel. This may be conceptually contrasted to the celebrated approximate deconvolution approaches where a filter shape is predefined for an iterative deconvolution process. We demonstrate that the proposed blind deconvolution network performs exceptionally well in the a-priori testing of both two-dimensional Kraichnan and three-dimensional Kolmogorov turbulence and shows promise in forming the backbone of a physics-augmented data-driven closure for the Navier-Stokes equations.

Real-space mapping of topological invariants using artificial neural networks

NASA Astrophysics Data System (ADS)

Carvalho, D.; García-Martínez, N. A.; Lado, J. L.; Fernández-Rossier, J.

2018-03-01

Topological invariants allow one to characterize Hamiltonians, predicting the existence of topologically protected in-gap modes. Those invariants can be computed by tracing the evolution of the occupied wave functions under twisted boundary conditions. However, those procedures do not allow one to calculate a topological invariant by evaluating the system locally, and thus require information about the wave functions in the whole system. Here we show that artificial neural networks can be trained to identify the topological order by evaluating a local projection of the density matrix. We demonstrate this for two different models, a one-dimensional topological superconductor and a two-dimensional quantum anomalous Hall state, both with spatially modulated parameters. Our neural network correctly identifies the different topological domains in real space, predicting the location of in-gap states. By combining a neural network with a calculation of the electronic states that uses the kernel polynomial method, we show that the local evaluation of the invariant can be carried out by evaluating a local quantity, in particular for systems without translational symmetry consisting of tens of thousands of atoms. Our results show that supervised learning is an efficient methodology to characterize the local topology of a system.

Using deep learning in image hyper spectral segmentation, classification, and detection

NASA Astrophysics Data System (ADS)

Zhao, Xiuying; Su, Zhenyu

2018-02-01

Recent years have shown that deep learning neural networks are a valuable tool in the field of computer vision. Deep learning method can be used in applications like remote sensing such as Land cover Classification, Detection of Vehicle in Satellite Images, Hyper spectral Image classification. This paper addresses the use of the deep learning artificial neural network in Satellite image segmentation. Image segmentation plays an important role in image processing. The hue of the remote sensing image often has a large hue difference, which will result in the poor display of the images in the VR environment. Image segmentation is a pre processing technique applied to the original images and splits the image into many parts which have different hue to unify the color. Several computational models based on supervised, unsupervised, parametric, probabilistic region based image segmentation techniques have been proposed. Recently, one of the machine learning technique known as, deep learning with convolution neural network has been widely used for development of efficient and automatic image segmentation models. In this paper, we focus on study of deep neural convolution network and its variants for automatic image segmentation rather than traditional image segmentation strategies.

Fuzzy recognition of noncompact musical objects

NASA Astrophysics Data System (ADS)

Cristobal Salas, Alfredo; Tchernykh, Andrei

1997-03-01

This article describes and compares some techniques to extract attributes from black and white images which contain musical objects. The inertia moment, the central moments and the wavelet transform methods are used to describe the images. Two supervised neural networks are applied to classify the images: backpropagation and fuzzy backpropagation. The results are compared.

Supervised Learning Applied to Air Traffic Trajectory Classification

NASA Technical Reports Server (NTRS)

Bosson, Christabelle S.; Nikoleris, Tasos

2018-01-01

Given the recent increase of interest in introducing new vehicle types and missions into the National Airspace System, a transition towards a more autonomous air traffic control system is required in order to enable and handle increased density and complexity. This paper presents an exploratory effort of the needed autonomous capabilities by exploring supervised learning techniques in the context of aircraft trajectories. In particular, it focuses on the application of machine learning algorithms and neural network models to a runway recognition trajectory-classification study. It investigates the applicability and effectiveness of various classifiers using datasets containing trajectory records for a month of air traffic. A feature importance and sensitivity analysis are conducted to challenge the chosen time-based datasets and the ten selected features. The study demonstrates that classification accuracy levels of 90% and above can be reached in less than 40 seconds of training for most machine learning classifiers when one track data point, described by the ten selected features at a particular time step, per trajectory is used as input. It also shows that neural network models can achieve similar accuracy levels but at higher training time costs.

Word embeddings and recurrent neural networks based on Long-Short Term Memory nodes in supervised biomedical word sense disambiguation.

PubMed

Jimeno Yepes, Antonio

2017-09-01

Word sense disambiguation helps identifying the proper sense of ambiguous words in text. With large terminologies such as the UMLS Metathesaurus ambiguities appear and highly effective disambiguation methods are required. Supervised learning algorithm methods are used as one of the approaches to perform disambiguation. Features extracted from the context of an ambiguous word are used to identify the proper sense of such a word. The type of features have an impact on machine learning methods, thus affect disambiguation performance. In this work, we have evaluated several types of features derived from the context of the ambiguous word and we have explored as well more global features derived from MEDLINE using word embeddings. Results show that word embeddings improve the performance of more traditional features and allow as well using recurrent neural network classifiers based on Long-Short Term Memory (LSTM) nodes. The combination of unigrams and word embeddings with an SVM sets a new state of the art performance with a macro accuracy of 95.97 in the MSH WSD data set. Copyright © 2017 Elsevier Inc. All rights reserved.

Elucidation and short-term forecasting of microcystin concentrations in Lake Suwa (Japan) by means of artificial neural networks and evolutionary algorithms.

PubMed

Chan, Wai Sum; Recknagel, Friedrich; Cao, Hongqing; Park, Ho-Dong

2007-05-01

Non-supervised artificial neural networks (ANN) and hybrid evolutionary algorithms (EA) were applied to analyse and model 12 years of limnological time-series data of the shallow hypertrophic Lake Suwa in Japan. The results have improved understanding of relationships between changing microcystin concentrations, Microcystis species abundances and annual rainfall intensity. The data analysis by non-supervised ANN revealed that total Microcystis abundance and extra-cellular microcystin concentrations in typical dry years are much higher than those in typical wet years. It also showed that high microcystin concentrations in dry years coincided with the dominance of the toxic Microcystis viridis whilst in typical wet years non-toxic Microcystis ichthyoblabe were dominant. Hybrid EA were used to discover rule sets to explain and forecast the occurrence of high microcystin concentrations in relation to water quality and climate conditions. The results facilitated early warning by 3-days-ahead forecasting of microcystin concentrations based on limnological and meteorological input data, achieving an r(2)=0.74 for testing.

ChemNet: A Transferable and Generalizable Deep Neural Network for Small-Molecule Property Prediction

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

Goh, Garrett B.; Siegel, Charles M.; Vishnu, Abhinav

With access to large datasets, deep neural networks through representation learning have been able to identify patterns from raw data, achieving human-level accuracy in image and speech recognition tasks. However, in chemistry, availability of large standardized and labelled datasets is scarce, and with a multitude of chemical properties of interest, chemical data is inherently small and fragmented. In this work, we explore transfer learning techniques in conjunction with the existing Chemception CNN model, to create a transferable and generalizable deep neural network for small-molecule property prediction. Our latest model, ChemNet learns in a semi-supervised manner from inexpensive labels computed frommore » the ChEMBL database. When fine-tuned to the Tox21, HIV and FreeSolv dataset, which are 3 separate chemical tasks that ChemNet was not originally trained on, we demonstrate that ChemNet exceeds the performance of existing Chemception models, contemporary MLP models that trains on molecular fingerprints, and it matches the performance of the ConvGraph algorithm, the current state-of-the-art. Furthermore, as ChemNet has been pre-trained on a large diverse chemical database, it can be used as a universal “plug-and-play” deep neural network, which accelerates the deployment of deep neural networks for the prediction of novel small-molecule chemical properties.« less

Mastering the game of Go with deep neural networks and tree search.

PubMed

Silver, David; Huang, Aja; Maddison, Chris J; Guez, Arthur; Sifre, Laurent; van den Driessche, George; Schrittwieser, Julian; Antonoglou, Ioannis; Panneershelvam, Veda; Lanctot, Marc; Dieleman, Sander; Grewe, Dominik; Nham, John; Kalchbrenner, Nal; Sutskever, Ilya; Lillicrap, Timothy; Leach, Madeleine; Kavukcuoglu, Koray; Graepel, Thore; Hassabis, Demis

2016-01-28

The game of Go has long been viewed as the most challenging of classic games for artificial intelligence owing to its enormous search space and the difficulty of evaluating board positions and moves. Here we introduce a new approach to computer Go that uses 'value networks' to evaluate board positions and 'policy networks' to select moves. These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play. Without any lookahead search, the neural networks play Go at the level of state-of-the-art Monte Carlo tree search programs that simulate thousands of random games of self-play. We also introduce a new search algorithm that combines Monte Carlo simulation with value and policy networks. Using this search algorithm, our program AlphaGo achieved a 99.8% winning rate against other Go programs, and defeated the human European Go champion by 5 games to 0. This is the first time that a computer program has defeated a human professional player in the full-sized game of Go, a feat previously thought to be at least a decade away.

Models of Acetylcholine and Dopamine Signals Differentially Improve Neural Representations

PubMed Central

Holca-Lamarre, Raphaël; Lücke, Jörg; Obermayer, Klaus

2017-01-01

Biological and artificial neural networks (ANNs) represent input signals as patterns of neural activity. In biology, neuromodulators can trigger important reorganizations of these neural representations. For instance, pairing a stimulus with the release of either acetylcholine (ACh) or dopamine (DA) evokes long lasting increases in the responses of neurons to the paired stimulus. The functional roles of ACh and DA in rearranging representations remain largely unknown. Here, we address this question using a Hebbian-learning neural network model. Our aim is both to gain a functional understanding of ACh and DA transmission in shaping biological representations and to explore neuromodulator-inspired learning rules for ANNs. We model the effects of ACh and DA on synaptic plasticity and confirm that stimuli coinciding with greater neuromodulator activation are over represented in the network. We then simulate the physiological release schedules of ACh and DA. We measure the impact of neuromodulator release on the network's representation and on its performance on a classification task. We find that ACh and DA trigger distinct changes in neural representations that both improve performance. The putative ACh signal redistributes neural preferences so that more neurons encode stimulus classes that are challenging for the network. The putative DA signal adapts synaptic weights so that they better match the classes of the task at hand. Our model thus offers a functional explanation for the effects of ACh and DA on cortical representations. Additionally, our learning algorithm yields performances comparable to those of state-of-the-art optimisation methods in multi-layer perceptrons while requiring weaker supervision signals and interacting with synaptically-local weight updates. PMID:28690509

Experimental Verification of Electric Drive Technologies Based on Artificial Intelligence Tools

NASA Technical Reports Server (NTRS)

Rubaai, Ahmed; Ricketts, Daniel; Kotaru, Raj; Thomas, Robert; Noga, Donald F. (Technical Monitor); Kankam, Mark D. (Technical Monitor)

2000-01-01

In this report, a fully integrated prototype of a flight servo control system is successfully developed and implemented using brushless dc motors. The control system is developed by the fuzzy logic theory, and implemented with a multilayer neural network. First, a neural network-based architecture is introduced for fuzzy logic control. The characteristic rules and their membership functions of fuzzy systems are represented as the processing nodes in the neural network structure. The network structure and the parameter learning are performed simultaneously and online in the fuzzy-neural network system. The structure learning is based on the partition of input space. The parameter learning is based on the supervised gradient decent method, using a delta adaptation law. Using experimental setup, the performance of the proposed control system is evaluated under various operating conditions. Test results are presented and discussed in the report. The proposed learning control system has several advantages, namely, simple structure and learning capability, robustness and high tracking performance and few nodes at hidden layers. In comparison with the PI controller, the proposed fuzzy-neural network system can yield a better dynamic performance with shorter settling time, and without overshoot. Experimental results have shown that the proposed control system is adaptive and robust in responding to a wide range of operating conditions. In summary, the goal of this study is to design and implement-advanced servosystems to actuate control surfaces for flight vehicles, namely, aircraft and helicopters, missiles and interceptors, and mini- and micro-air vehicles.

Accurate Natural Trail Detection Using a Combination of a Deep Neural Network and Dynamic Programming.

PubMed

Adhikari, Shyam Prasad; Yang, Changju; Slot, Krzysztof; Kim, Hyongsuk

2018-01-10

This paper presents a vision sensor-based solution to the challenging problem of detecting and following trails in highly unstructured natural environments like forests, rural areas and mountains, using a combination of a deep neural network and dynamic programming. The deep neural network (DNN) concept has recently emerged as a very effective tool for processing vision sensor signals. A patch-based DNN is trained with supervised data to classify fixed-size image patches into "trail" and "non-trail" categories, and reshaped to a fully convolutional architecture to produce trail segmentation map for arbitrary-sized input images. As trail and non-trail patches do not exhibit clearly defined shapes or forms, the patch-based classifier is prone to misclassification, and produces sub-optimal trail segmentation maps. Dynamic programming is introduced to find an optimal trail on the sub-optimal DNN output map. Experimental results showing accurate trail detection for real-world trail datasets captured with a head mounted vision system are presented.

Robust model predictive control of nonlinear systems with unmodeled dynamics and bounded uncertainties based on neural networks.

PubMed

Yan, Zheng; Wang, Jun

2014-03-01

This paper presents a neural network approach to robust model predictive control (MPC) for constrained discrete-time nonlinear systems with unmodeled dynamics affected by bounded uncertainties. The exact nonlinear model of underlying process is not precisely known, but a partially known nominal model is available. This partially known nonlinear model is first decomposed to an affine term plus an unknown high-order term via Jacobian linearization. The linearization residue combined with unmodeled dynamics is then modeled using an extreme learning machine via supervised learning. The minimax methodology is exploited to deal with bounded uncertainties. The minimax optimization problem is reformulated as a convex minimization problem and is iteratively solved by a two-layer recurrent neural network. The proposed neurodynamic approach to nonlinear MPC improves the computational efficiency and sheds a light for real-time implementability of MPC technology. Simulation results are provided to substantiate the effectiveness and characteristics of the proposed approach.

Stochastic associative memory

NASA Astrophysics Data System (ADS)

Baumann, Erwin W.; Williams, David L.

1993-08-01

Artificial neural networks capable of learning and recalling stochastic associations between non-deterministic quantities have received relatively little attention to date. One potential application of such stochastic associative networks is the generation of sensory 'expectations' based on arbitrary subsets of sensor inputs to support anticipatory and investigate behavior in sensor-based robots. Another application of this type of associative memory is the prediction of how a scene will look in one spectral band, including noise, based upon its appearance in several other wavebands. This paper describes a semi-supervised neural network architecture composed of self-organizing maps associated through stochastic inter-layer connections. This 'Stochastic Associative Memory' (SAM) can learn and recall non-deterministic associations between multi-dimensional probability density functions. The stochastic nature of the network also enables it to represent noise distributions that are inherent in any true sensing process. The SAM architecture, training process, and initial application to sensor image prediction are described. Relationships to Fuzzy Associative Memory (FAM) are discussed.

Learning and optimization with cascaded VLSI neural network building-block chips

NASA Technical Reports Server (NTRS)

Duong, T.; Eberhardt, S. P.; Tran, M.; Daud, T.; Thakoor, A. P.

1992-01-01

To demonstrate the versatility of the building-block approach, two neural network applications were implemented on cascaded analog VLSI chips. Weights were implemented using 7-b multiplying digital-to-analog converter (MDAC) synapse circuits, with 31 x 32 and 32 x 32 synapses per chip. A novel learning algorithm compatible with analog VLSI was applied to the two-input parity problem. The algorithm combines dynamically evolving architecture with limited gradient-descent backpropagation for efficient and versatile supervised learning. To implement the learning algorithm in hardware, synapse circuits were paralleled for additional quantization levels. The hardware-in-the-loop learning system allocated 2-5 hidden neurons for parity problems. Also, a 7 x 7 assignment problem was mapped onto a cascaded 64-neuron fully connected feedback network. In 100 randomly selected problems, the network found optimal or good solutions in most cases, with settling times in the range of 7-100 microseconds.

Ensemble learning in fixed expansion layer networks for mitigating catastrophic forgetting.

PubMed

Coop, Robert; Mishtal, Aaron; Arel, Itamar

2013-10-01

Catastrophic forgetting is a well-studied attribute of most parameterized supervised learning systems. A variation of this phenomenon, in the context of feedforward neural networks, arises when nonstationary inputs lead to loss of previously learned mappings. The majority of the schemes proposed in the literature for mitigating catastrophic forgetting were not data driven and did not scale well. We introduce the fixed expansion layer (FEL) feedforward neural network, which embeds a sparsely encoding hidden layer to help mitigate forgetting of prior learned representations. In addition, we investigate a novel framework for training ensembles of FEL networks, based on exploiting an information-theoretic measure of diversity between FEL learners, to further control undesired plasticity. The proposed methodology is demonstrated on a basic classification task, clearly emphasizing its advantages over existing techniques. The architecture proposed can be enhanced to address a range of computational intelligence tasks, such as regression problems and system control.

Supervised Learning Based on Temporal Coding in Spiking Neural Networks.

PubMed

Mostafa, Hesham

2017-08-01

Gradient descent training techniques are remarkably successful in training analog-valued artificial neural networks (ANNs). Such training techniques, however, do not transfer easily to spiking networks due to the spike generation hard nonlinearity and the discrete nature of spike communication. We show that in a feedforward spiking network that uses a temporal coding scheme where information is encoded in spike times instead of spike rates, the network input-output relation is differentiable almost everywhere. Moreover, this relation is piecewise linear after a transformation of variables. Methods for training ANNs thus carry directly to the training of such spiking networks as we show when training on the permutation invariant MNIST task. In contrast to rate-based spiking networks that are often used to approximate the behavior of ANNs, the networks we present spike much more sparsely and their behavior cannot be directly approximated by conventional ANNs. Our results highlight a new approach for controlling the behavior of spiking networks with realistic temporal dynamics, opening up the potential for using these networks to process spike patterns with complex temporal information.

Representational Distance Learning for Deep Neural Networks

PubMed Central

McClure, Patrick; Kriegeskorte, Nikolaus

2016-01-01

Deep neural networks (DNNs) provide useful models of visual representational transformations. We present a method that enables a DNN (student) to learn from the internal representational spaces of a reference model (teacher), which could be another DNN or, in the future, a biological brain. Representational spaces of the student and the teacher are characterized by representational distance matrices (RDMs). We propose representational distance learning (RDL), a stochastic gradient descent method that drives the RDMs of the student to approximate the RDMs of the teacher. We demonstrate that RDL is competitive with other transfer learning techniques for two publicly available benchmark computer vision datasets (MNIST and CIFAR-100), while allowing for architectural differences between student and teacher. By pulling the student's RDMs toward those of the teacher, RDL significantly improved visual classification performance when compared to baseline networks that did not use transfer learning. In the future, RDL may enable combined supervised training of deep neural networks using task constraints (e.g., images and category labels) and constraints from brain-activity measurements, so as to build models that replicate the internal representational spaces of biological brains. PMID:28082889

Representational Distance Learning for Deep Neural Networks.

PubMed

McClure, Patrick; Kriegeskorte, Nikolaus

2016-01-01

Deep neural networks (DNNs) provide useful models of visual representational transformations. We present a method that enables a DNN (student) to learn from the internal representational spaces of a reference model (teacher), which could be another DNN or, in the future, a biological brain. Representational spaces of the student and the teacher are characterized by representational distance matrices (RDMs). We propose representational distance learning (RDL), a stochastic gradient descent method that drives the RDMs of the student to approximate the RDMs of the teacher. We demonstrate that RDL is competitive with other transfer learning techniques for two publicly available benchmark computer vision datasets (MNIST and CIFAR-100), while allowing for architectural differences between student and teacher. By pulling the student's RDMs toward those of the teacher, RDL significantly improved visual classification performance when compared to baseline networks that did not use transfer learning. In the future, RDL may enable combined supervised training of deep neural networks using task constraints (e.g., images and category labels) and constraints from brain-activity measurements, so as to build models that replicate the internal representational spaces of biological brains.

PSF estimation for defocus blurred image based on quantum back-propagation neural network

NASA Astrophysics Data System (ADS)

Gao, Kun; Zhang, Yan; Shao, Xiao-guang; Liu, Ying-hui; Ni, Guoqiang

2010-11-01

Images obtained by an aberration-free system are defocused blur due to motion in depth and/or zooming. The precondition of restoring the degraded image is to estimate point spread function (PSF) of the imaging system as precisely as possible. But it is difficult to identify the analytic model of PSF precisely due to the complexity of the degradation process. Inspired by the similarity between the quantum process and imaging process in the probability and statistics fields, one reformed multilayer quantum neural network (QNN) is proposed to estimate PSF of the defocus blurred image. Different from the conventional artificial neural network (ANN), an improved quantum neuron model is used in the hidden layer instead, which introduces a 2-bit controlled NOT quantum gate to control output and adopts 2 texture and edge features as the input vectors. The supervised back-propagation learning rule is adopted to train network based on training sets from the historical images. Test results show that this method owns excellent features of high precision and strong generalization ability.

Chaotic map clustering algorithm for EEG analysis

NASA Astrophysics Data System (ADS)

Bellotti, R.; De Carlo, F.; Stramaglia, S.

2004-03-01

The non-parametric chaotic map clustering algorithm has been applied to the analysis of electroencephalographic signals, in order to recognize the Huntington's disease, one of the most dangerous pathologies of the central nervous system. The performance of the method has been compared with those obtained through parametric algorithms, as K-means and deterministic annealing, and supervised multi-layer perceptron. While supervised neural networks need a training phase, performed by means of data tagged by the genetic test, and the parametric methods require a prior choice of the number of classes to find, the chaotic map clustering gives a natural evidence of the pathological class, without any training or supervision, thus providing a new efficient methodology for the recognition of patterns affected by the Huntington's disease.

Retrieving Tract Variables From Acoustics: A Comparison of Different Machine Learning Strategies.

PubMed

Mitra, Vikramjit; Nam, Hosung; Espy-Wilson, Carol Y; Saltzman, Elliot; Goldstein, Louis

2010-09-13

Many different studies have claimed that articulatory information can be used to improve the performance of automatic speech recognition systems. Unfortunately, such articulatory information is not readily available in typical speaker-listener situations. Consequently, such information has to be estimated from the acoustic signal in a process which is usually termed "speech-inversion." This study aims to propose and compare various machine learning strategies for speech inversion: Trajectory mixture density networks (TMDNs), feedforward artificial neural networks (FF-ANN), support vector regression (SVR), autoregressive artificial neural network (AR-ANN), and distal supervised learning (DSL). Further, using a database generated by the Haskins Laboratories speech production model, we test the claim that information regarding constrictions produced by the distinct organs of the vocal tract (vocal tract variables) is superior to flesh-point information (articulatory pellet trajectories) for the inversion process.

Software for hyperspectral, joint photographic experts group (.JPG), portable network graphics (.PNG) and tagged image file format (.TIFF) segmentation

NASA Astrophysics Data System (ADS)

Bruno, L. S.; Rodrigo, B. P.; Lucio, A. de C. Jorge

2016-10-01

This paper presents a system developed by an application of a neural network Multilayer Perceptron for drone acquired agricultural image segmentation. This application allows a supervised user training the classes that will posteriorly be interpreted by neural network. These classes will be generated manually with pre-selected attributes in the application. After the attribute selection a segmentation process is made to allow the relevant information extraction for different types of images, RGB or Hyperspectral. The application allows extracting the geographical coordinates from the image metadata, geo referencing all pixels on the image. In spite of excessive memory consume on hyperspectral images regions of interest, is possible to perform segmentation, using bands chosen by user that can be combined in different ways to obtain different results.

Learning a trajectory using adjoint functions and teacher forcing

NASA Technical Reports Server (NTRS)

Toomarian, Nikzad B.; Barhen, Jacob

1992-01-01

A new methodology for faster supervised temporal learning in nonlinear neural networks is presented which builds upon the concept of adjoint operators to allow fast computation of the gradients of an error functional with respect to all parameters of the neural architecture, and exploits the concept of teacher forcing to incorporate information on the desired output into the activation dynamics. The importance of the initial or final time conditions for the adjoint equations is discussed. A new algorithm is presented in which the adjoint equations are solved simultaneously (i.e., forward in time) with the activation dynamics of the neural network. We also indicate how teacher forcing can be modulated in time as learning proceeds. The results obtained show that the learning time is reduced by one to two orders of magnitude with respect to previously published results, while trajectory tracking is significantly improved. The proposed methodology makes hardware implementation of temporal learning attractive for real-time applications.

The probabilistic neural network architecture for high speed classification of remotely sensed imagery

NASA Technical Reports Server (NTRS)

Chettri, Samir R.; Cromp, Robert F.

1993-01-01

In this paper we discuss a neural network architecture (the Probabilistic Neural Net or the PNN) that, to the best of our knowledge, has not previously been applied to remotely sensed data. The PNN is a supervised non-parametric classification algorithm as opposed to the Gaussian maximum likelihood classifier (GMLC). The PNN works by fitting a Gaussian kernel to each training point. The width of the Gaussian is controlled by a tuning parameter called the window width. If very small widths are used, the method is equivalent to the nearest neighbor method. For large windows, the PNN behaves like the GMLC. The basic implementation of the PNN requires no training time at all. In this respect it is far better than the commonly used backpropagation neural network which can be shown to take O(N6) time for training where N is the dimensionality of the input vector. In addition the PNN can be implemented in a feed forward mode in hardware. The disadvantage of the PNN is that it requires all the training data to be stored. Some solutions to this problem are discussed in the paper. Finally, we discuss the accuracy of the PNN with respect to the GMLC and the backpropagation neural network (BPNN). The PNN is shown to be better than GMLC and not as good as the BPNN with regards to classification accuracy.

Identification and discrimination of oral asaccharolytic Eubacterium spp. by pyrolysis mass spectrometry and artificial neural networks.

PubMed

Goodacre, R; Hiom, S J; Cheeseman, S L; Murdoch, D; Weightman, A J; Wade, W G

1996-02-01

Curie-point pyrolysis mass spectra were obtained from 29 oral asaccharolytic Eubacterium strains and 6 abscess isolates previously identified as Peptostreptococcus heliotrinreducens. Pyrolysis mass spectrometry (PyMS) with cluster analysis was able to clarify the taxonomic position of this group of organisms. Artificial neural networks (ANNS) were then trained by supervised learning (with the back-propagation algorithm) to recognize the strains from their pyrolysis mass spectra; all Eubacterium strains were correctly identified, and the abscess isolates were identified as un-named Eubacterium taxon C2 and were distinct from the type strain of P. heliotrinreducens. These results demonstrate that the combination of PyMS and ANNs provides a rapid and accurate identification technique.

Detecting event-related changes in organizational networks using optimized neural network models.

PubMed

Li, Ze; Sun, Duoyong; Zhu, Renqi; Lin, Zihan

2017-01-01

Organizational external behavior changes are caused by the internal structure and interactions. External behaviors are also known as the behavioral events of an organization. Detecting event-related changes in organizational networks could efficiently be used to monitor the dynamics of organizational behaviors. Although many different methods have been used to detect changes in organizational networks, these methods usually ignore the correlation between the internal structure and external events. Event-related change detection considers the correlation and could be used for event recognition based on social network modeling and supervised classification. Detecting event-related changes could be effectively useful in providing early warnings and faster responses to both positive and negative organizational activities. In this study, event-related change in an organizational network was defined, and artificial neural network models were used to quantitatively determine whether and when a change occurred. To achieve a higher accuracy, Back Propagation Neural Networks (BPNNs) were optimized using Genetic Algorithms (GAs) and Particle Swarm Optimization (PSO). We showed the feasibility of the proposed method by comparing its performance with that of other methods using two cases. The results suggested that the proposed method could identify organizational events based on a correlation between the organizational networks and events. The results also suggested that the proposed method not only has a higher precision but also has a better robustness than the previously used techniques.

Detecting event-related changes in organizational networks using optimized neural network models

PubMed Central

Sun, Duoyong; Zhu, Renqi; Lin, Zihan

2017-01-01

Organizational external behavior changes are caused by the internal structure and interactions. External behaviors are also known as the behavioral events of an organization. Detecting event-related changes in organizational networks could efficiently be used to monitor the dynamics of organizational behaviors. Although many different methods have been used to detect changes in organizational networks, these methods usually ignore the correlation between the internal structure and external events. Event-related change detection considers the correlation and could be used for event recognition based on social network modeling and supervised classification. Detecting event-related changes could be effectively useful in providing early warnings and faster responses to both positive and negative organizational activities. In this study, event-related change in an organizational network was defined, and artificial neural network models were used to quantitatively determine whether and when a change occurred. To achieve a higher accuracy, Back Propagation Neural Networks (BPNNs) were optimized using Genetic Algorithms (GAs) and Particle Swarm Optimization (PSO). We showed the feasibility of the proposed method by comparing its performance with that of other methods using two cases. The results suggested that the proposed method could identify organizational events based on a correlation between the organizational networks and events. The results also suggested that the proposed method not only has a higher precision but also has a better robustness than the previously used techniques. PMID:29190799

Deep learning in the small sample size setting: cascaded feed forward neural networks for medical image segmentation

NASA Astrophysics Data System (ADS)

Gaonkar, Bilwaj; Hovda, David; Martin, Neil; Macyszyn, Luke

2016-03-01

Deep Learning, refers to large set of neural network based algorithms, have emerged as promising machine- learning tools in the general imaging and computer vision domains. Convolutional neural networks (CNNs), a specific class of deep learning algorithms, have been extremely effective in object recognition and localization in natural images. A characteristic feature of CNNs, is the use of a locally connected multi layer topology that is inspired by the animal visual cortex (the most powerful vision system in existence). While CNNs, perform admirably in object identification and localization tasks, typically require training on extremely large datasets. Unfortunately, in medical image analysis, large datasets are either unavailable or are extremely expensive to obtain. Further, the primary tasks in medical imaging are organ identification and segmentation from 3D scans, which are different from the standard computer vision tasks of object recognition. Thus, in order to translate the advantages of deep learning to medical image analysis, there is a need to develop deep network topologies and training methodologies, that are geared towards medical imaging related tasks and can work in a setting where dataset sizes are relatively small. In this paper, we present a technique for stacked supervised training of deep feed forward neural networks for segmenting organs from medical scans. Each `neural network layer' in the stack is trained to identify a sub region of the original image, that contains the organ of interest. By layering several such stacks together a very deep neural network is constructed. Such a network can be used to identify extremely small regions of interest in extremely large images, inspite of a lack of clear contrast in the signal or easily identifiable shape characteristics. What is even more intriguing is that the network stack achieves accurate segmentation even when it is trained on a single image with manually labelled ground truth. We validate this approach,using a publicly available head and neck CT dataset. We also show that a deep neural network of similar depth, if trained directly using backpropagation, cannot acheive the tasks achieved using our layer wise training paradigm.

Multi-level deep supervised networks for retinal vessel segmentation.

PubMed

Mo, Juan; Zhang, Lei

2017-12-01

Changes in the appearance of retinal blood vessels are an important indicator for various ophthalmologic and cardiovascular diseases, including diabetes, hypertension, arteriosclerosis, and choroidal neovascularization. Vessel segmentation from retinal images is very challenging because of low blood vessel contrast, intricate vessel topology, and the presence of pathologies such as microaneurysms and hemorrhages. To overcome these challenges, we propose a neural network-based method for vessel segmentation. A deep supervised fully convolutional network is developed by leveraging multi-level hierarchical features of the deep networks. To improve the discriminative capability of features in lower layers of the deep network and guide the gradient back propagation to overcome gradient vanishing, deep supervision with auxiliary classifiers is incorporated in some intermediate layers of the network. Moreover, the transferred knowledge learned from other domains is used to alleviate the issue of insufficient medical training data. The proposed approach does not rely on hand-crafted features and needs no problem-specific preprocessing or postprocessing, which reduces the impact of subjective factors. We evaluate the proposed method on three publicly available databases, the DRIVE, STARE, and CHASE_DB1 databases. Extensive experiments demonstrate that our approach achieves better or comparable performance to state-of-the-art methods with a much faster processing speed, making it suitable for real-world clinical applications. The results of cross-training experiments demonstrate its robustness with respect to the training set. The proposed approach segments retinal vessels accurately with a much faster processing speed and can be easily applied to other biomedical segmentation tasks.

A classifier neural network for rotordynamic systems

NASA Astrophysics Data System (ADS)

Ganesan, R.; Jionghua, Jin; Sankar, T. S.

1995-07-01

A feedforward backpropagation neural network is formed to identify the stability characteristic of a high speed rotordynamic system. The principal focus resides in accounting for the instability due to the bearing clearance effects. The abnormal operating condition of 'normal-loose' Coulomb rub, that arises in units supported by hydrodynamic bearings or rolling element bearings, is analysed in detail. The multiple-parameter stability problem is formulated and converted to a set of three-parameter algebraic inequality equations. These three parameters map the wider range of physical parameters of commonly-used rotordynamic systems into a narrow closed region, that is used in the supervised learning of the neural network. A binary-type state of the system is expressed through these inequalities that are deduced from the analytical simulation of the rotor system. Both the hidden layer as well as functional-link networks are formed and the superiority of the functional-link network is established. Considering the real time interpretation and control of the rotordynamic system, the network reliability and the learning time are used as the evaluation criteria to assess the superiority of the functional-link network. This functional-link network is further trained using the parameter values of selected rotor systems, and the classifier network is formed. The success rate of stability status identification is obtained to assess the potentials of this classifier network. The classifier network is shown that it can also be used, for control purposes, as an 'advisory' system that suggests the optimum way of parameter adjustment.

An adaptive neural swarm approach for intrusion defense in ad hoc networks

NASA Astrophysics Data System (ADS)

Cannady, James

2011-06-01

Wireless sensor networks (WSN) and mobile ad hoc networks (MANET) are being increasingly deployed in critical applications due to the flexibility and extensibility of the technology. While these networks possess numerous advantages over traditional wireless systems in dynamic environments they are still vulnerable to many of the same types of host-based and distributed attacks common to those systems. Unfortunately, the limited power and bandwidth available in WSNs and MANETs, combined with the dynamic connectivity that is a defining characteristic of the technology, makes it extremely difficult to utilize traditional intrusion detection techniques. This paper describes an approach to accurately and efficiently detect potentially damaging activity in WSNs and MANETs. It enables the network as a whole to recognize attacks, anomalies, and potential vulnerabilities in a distributive manner that reflects the autonomic processes of biological systems. Each component of the network recognizes activity in its local environment and then contributes to the overall situational awareness of the entire system. The approach utilizes agent-based swarm intelligence to adaptively identify potential data sources on each node and on adjacent nodes throughout the network. The swarm agents then self-organize into modular neural networks that utilize a reinforcement learning algorithm to identify relevant behavior patterns in the data without supervision. Once the modular neural networks have established interconnectivity both locally and with neighboring nodes the analysis of events within the network can be conducted collectively in real-time. The approach has been shown to be extremely effective in identifying distributed network attacks.

Application of artificial neural network to fMRI regression analysis.

PubMed

Misaki, Masaya; Miyauchi, Satoru

2006-01-15

We used an artificial neural network (ANN) to detect correlations between event sequences and fMRI (functional magnetic resonance imaging) signals. The layered feed-forward neural network, given a series of events as inputs and the fMRI signal as a supervised signal, performed a non-linear regression analysis. This type of ANN is capable of approximating any continuous function, and thus this analysis method can detect any fMRI signals that correlated with corresponding events. Because of the flexible nature of ANNs, fitting to autocorrelation noise is a problem in fMRI analyses. We avoided this problem by using cross-validation and an early stopping procedure. The results showed that the ANN could detect various responses with different time courses. The simulation analysis also indicated an additional advantage of ANN over non-parametric methods in detecting parametrically modulated responses, i.e., it can detect various types of parametric modulations without a priori assumptions. The ANN regression analysis is therefore beneficial for exploratory fMRI analyses in detecting continuous changes in responses modulated by changes in input values.

Solving graph data issues using a layered architecture approach with applications to web spam detection.

PubMed

Scarselli, Franco; Tsoi, Ah Chung; Hagenbuchner, Markus; Noi, Lucia Di

2013-12-01

This paper proposes the combination of two state-of-the-art algorithms for processing graph input data, viz., the probabilistic mapping graph self organizing map, an unsupervised learning approach, and the graph neural network, a supervised learning approach. We organize these two algorithms in a cascade architecture containing a probabilistic mapping graph self organizing map, and a graph neural network. We show that this combined approach helps us to limit the long-term dependency problem that exists when training the graph neural network resulting in an overall improvement in performance. This is demonstrated in an application to a benchmark problem requiring the detection of spam in a relatively large set of web sites. It is found that the proposed method produces results which reach the state of the art when compared with some of the best results obtained by others using quite different approaches. A particular strength of our method is its applicability towards any input domain which can be represented as a graph. Copyright © 2013 Elsevier Ltd. All rights reserved.

Shakeout: A New Approach to Regularized Deep Neural Network Training.

PubMed

Kang, Guoliang; Li, Jun; Tao, Dacheng

2018-05-01

Recent years have witnessed the success of deep neural networks in dealing with a plenty of practical problems. Dropout has played an essential role in many successful deep neural networks, by inducing regularization in the model training. In this paper, we present a new regularized training approach: Shakeout. Instead of randomly discarding units as Dropout does at the training stage, Shakeout randomly chooses to enhance or reverse each unit's contribution to the next layer. This minor modification of Dropout has the statistical trait: the regularizer induced by Shakeout adaptively combines , and regularization terms. Our classification experiments with representative deep architectures on image datasets MNIST, CIFAR-10 and ImageNet show that Shakeout deals with over-fitting effectively and outperforms Dropout. We empirically demonstrate that Shakeout leads to sparser weights under both unsupervised and supervised settings. Shakeout also leads to the grouping effect of the input units in a layer. Considering the weights in reflecting the importance of connections, Shakeout is superior to Dropout, which is valuable for the deep model compression. Moreover, we demonstrate that Shakeout can effectively reduce the instability of the training process of the deep architecture.

Smoothing of cost function leads to faster convergence of neural network learning

NASA Astrophysics Data System (ADS)

Xu, Li-Qun; Hall, Trevor J.

1994-03-01

One of the major problems in supervised learning of neural networks is the inevitable local minima inherent in the cost function f(W,D). This often makes classic gradient-descent-based learning algorithms that calculate the weight updates for each iteration according to (Delta) W(t) equals -(eta) (DOT)$DELwf(W,D) powerless. In this paper we describe a new strategy to solve this problem, which, adaptively, changes the learning rate and manipulates the gradient estimator simultaneously. The idea is to implicitly convert the local- minima-laden cost function f((DOT)) into a sequence of its smoothed versions {f(beta t)}Ttequals1, which, subject to the parameter (beta) t, bears less details at time t equals 1 and gradually more later on, the learning is actually performed on this sequence of functionals. The corresponding smoothed global minima obtained in this way, {Wt}Ttequals1, thus progressively approximate W-the desired global minimum. Experimental results on a nonconvex function minimization problem and a typical neural network learning task are given, analyses and discussions of some important issues are provided.

Mastering the game of Go with deep neural networks and tree search

NASA Astrophysics Data System (ADS)

Silver, David; Huang, Aja; Maddison, Chris J.; Guez, Arthur; Sifre, Laurent; van den Driessche, George; Schrittwieser, Julian; Antonoglou, Ioannis; Panneershelvam, Veda; Lanctot, Marc; Dieleman, Sander; Grewe, Dominik; Nham, John; Kalchbrenner, Nal; Sutskever, Ilya; Lillicrap, Timothy; Leach, Madeleine; Kavukcuoglu, Koray; Graepel, Thore; Hassabis, Demis

2016-01-01

The game of Go has long been viewed as the most challenging of classic games for artificial intelligence owing to its enormous search space and the difficulty of evaluating board positions and moves. Here we introduce a new approach to computer Go that uses ‘value networks’ to evaluate board positions and ‘policy networks’ to select moves. These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play. Without any lookahead search, the neural networks play Go at the level of state-of-the-art Monte Carlo tree search programs that simulate thousands of random games of self-play. We also introduce a new search algorithm that combines Monte Carlo simulation with value and policy networks. Using this search algorithm, our program AlphaGo achieved a 99.8% winning rate against other Go programs, and defeated the human European Go champion by 5 games to 0. This is the first time that a computer program has defeated a human professional player in the full-sized game of Go, a feat previously thought to be at least a decade away.

Stable architectures for deep neural networks

NASA Astrophysics Data System (ADS)

Haber, Eldad; Ruthotto, Lars

2018-01-01

Deep neural networks have become invaluable tools for supervised machine learning, e.g. classification of text or images. While often offering superior results over traditional techniques and successfully expressing complicated patterns in data, deep architectures are known to be challenging to design and train such that they generalize well to new data. Critical issues with deep architectures are numerical instabilities in derivative-based learning algorithms commonly called exploding or vanishing gradients. In this paper, we propose new forward propagation techniques inspired by systems of ordinary differential equations (ODE) that overcome this challenge and lead to well-posed learning problems for arbitrarily deep networks. The backbone of our approach is our interpretation of deep learning as a parameter estimation problem of nonlinear dynamical systems. Given this formulation, we analyze stability and well-posedness of deep learning and use this new understanding to develop new network architectures. We relate the exploding and vanishing gradient phenomenon to the stability of the discrete ODE and present several strategies for stabilizing deep learning for very deep networks. While our new architectures restrict the solution space, several numerical experiments show their competitiveness with state-of-the-art networks.

Classification of spontaneous EEG signals in migraine

NASA Astrophysics Data System (ADS)

Bellotti, R.; De Carlo, F.; de Tommaso, M.; Lucente, M.

2007-08-01

We set up a classification system able to detect patients affected by migraine without aura, through the analysis of their spontaneous EEG patterns. First, the signals are characterized by means of wavelet-based features, than a supervised neural network is used to classify the multichannel data. For the feature extraction, scale-dependent and scale-independent methods are considered with a variety of wavelet functions. Both the approaches provide very high and almost comparable classification performances. A complete separation of the two groups is obtained when the data are plotted in the plane spanned by two suitable neural outputs.

Training Feedforward Neural Networks Using Symbiotic Organisms Search Algorithm.

PubMed

Wu, Haizhou; Zhou, Yongquan; Luo, Qifang; Basset, Mohamed Abdel

2016-01-01

Symbiotic organisms search (SOS) is a new robust and powerful metaheuristic algorithm, which stimulates the symbiotic interaction strategies adopted by organisms to survive and propagate in the ecosystem. In the supervised learning area, it is a challenging task to present a satisfactory and efficient training algorithm for feedforward neural networks (FNNs). In this paper, SOS is employed as a new method for training FNNs. To investigate the performance of the aforementioned method, eight different datasets selected from the UCI machine learning repository are employed for experiment and the results are compared among seven metaheuristic algorithms. The results show that SOS performs better than other algorithms for training FNNs in terms of converging speed. It is also proven that an FNN trained by the method of SOS has better accuracy than most algorithms compared.

Improving EEG-Based Driver Fatigue Classification Using Sparse-Deep Belief Networks.

PubMed

Chai, Rifai; Ling, Sai Ho; San, Phyo Phyo; Naik, Ganesh R; Nguyen, Tuan N; Tran, Yvonne; Craig, Ashley; Nguyen, Hung T

2017-01-01

This paper presents an improvement of classification performance for electroencephalography (EEG)-based driver fatigue classification between fatigue and alert states with the data collected from 43 participants. The system employs autoregressive (AR) modeling as the features extraction algorithm, and sparse-deep belief networks (sparse-DBN) as the classification algorithm. Compared to other classifiers, sparse-DBN is a semi supervised learning method which combines unsupervised learning for modeling features in the pre-training layer and supervised learning for classification in the following layer. The sparsity in sparse-DBN is achieved with a regularization term that penalizes a deviation of the expected activation of hidden units from a fixed low-level prevents the network from overfitting and is able to learn low-level structures as well as high-level structures. For comparison, the artificial neural networks (ANN), Bayesian neural networks (BNN), and original deep belief networks (DBN) classifiers are used. The classification results show that using AR feature extractor and DBN classifiers, the classification performance achieves an improved classification performance with a of sensitivity of 90.8%, a specificity of 90.4%, an accuracy of 90.6%, and an area under the receiver operating curve (AUROC) of 0.94 compared to ANN (sensitivity at 80.8%, specificity at 77.8%, accuracy at 79.3% with AUC-ROC of 0.83) and BNN classifiers (sensitivity at 84.3%, specificity at 83%, accuracy at 83.6% with AUROC of 0.87). Using the sparse-DBN classifier, the classification performance improved further with sensitivity of 93.9%, a specificity of 92.3%, and an accuracy of 93.1% with AUROC of 0.96. Overall, the sparse-DBN classifier improved accuracy by 13.8, 9.5, and 2.5% over ANN, BNN, and DBN classifiers, respectively.

Improving EEG-Based Driver Fatigue Classification Using Sparse-Deep Belief Networks

PubMed Central

Chai, Rifai; Ling, Sai Ho; San, Phyo Phyo; Naik, Ganesh R.; Nguyen, Tuan N.; Tran, Yvonne; Craig, Ashley; Nguyen, Hung T.

2017-01-01

This paper presents an improvement of classification performance for electroencephalography (EEG)-based driver fatigue classification between fatigue and alert states with the data collected from 43 participants. The system employs autoregressive (AR) modeling as the features extraction algorithm, and sparse-deep belief networks (sparse-DBN) as the classification algorithm. Compared to other classifiers, sparse-DBN is a semi supervised learning method which combines unsupervised learning for modeling features in the pre-training layer and supervised learning for classification in the following layer. The sparsity in sparse-DBN is achieved with a regularization term that penalizes a deviation of the expected activation of hidden units from a fixed low-level prevents the network from overfitting and is able to learn low-level structures as well as high-level structures. For comparison, the artificial neural networks (ANN), Bayesian neural networks (BNN), and original deep belief networks (DBN) classifiers are used. The classification results show that using AR feature extractor and DBN classifiers, the classification performance achieves an improved classification performance with a of sensitivity of 90.8%, a specificity of 90.4%, an accuracy of 90.6%, and an area under the receiver operating curve (AUROC) of 0.94 compared to ANN (sensitivity at 80.8%, specificity at 77.8%, accuracy at 79.3% with AUC-ROC of 0.83) and BNN classifiers (sensitivity at 84.3%, specificity at 83%, accuracy at 83.6% with AUROC of 0.87). Using the sparse-DBN classifier, the classification performance improved further with sensitivity of 93.9%, a specificity of 92.3%, and an accuracy of 93.1% with AUROC of 0.96. Overall, the sparse-DBN classifier improved accuracy by 13.8, 9.5, and 2.5% over ANN, BNN, and DBN classifiers, respectively. PMID:28326009

Deep Neural Network Based Supervised Speech Segregation Generalizes to Novel Noises through Large-scale Training

DTIC Science & Technology

2015-01-01

Table 2: Segregation results in terms of STOI on a variety of novel noises (SNR=-2 dB) Babble-20 Cafeteria Factory Babble-100 Living Room Cafe Park...NOISEX-92 corpus [13], and a living room, a cafe and a park noise from the DEMAND corpus [12]. To put the performance of the noise-independent model in

Porosity Estimation By Artificial Neural Networks Inversion . Application to Algerian South Field

NASA Astrophysics Data System (ADS)

Eladj, Said; Aliouane, Leila; Ouadfeul, Sid-Ali

2017-04-01

One of the main geophysicist's current challenge is the discovery and the study of stratigraphic traps, this last is a difficult task and requires a very fine analysis of the seismic data. The seismic data inversion allows obtaining lithological and stratigraphic information for the reservoir characterization . However, when solving the inverse problem we encounter difficult problems such as: Non-existence and non-uniqueness of the solution add to this the instability of the processing algorithm. Therefore, uncertainties in the data and the non-linearity of the relationship between the data and the parameters must be taken seriously. In this case, the artificial intelligence techniques such as Artificial Neural Networks(ANN) is used to resolve this ambiguity, this can be done by integrating different physical properties data which requires a supervised learning methods. In this work, we invert the acoustic impedance 3D seismic cube using the colored inversion method, then, the introduction of the acoustic impedance volume resulting from the first step as an input of based model inversion method allows to calculate the Porosity volume using the Multilayer Perceptron Artificial Neural Network. Application to an Algerian South hydrocarbon field clearly demonstrate the power of the proposed processing technique to predict the porosity for seismic data, obtained results can be used for reserves estimation, permeability prediction, recovery factor and reservoir monitoring. Keywords: Artificial Neural Networks, inversion, non-uniqueness , nonlinear, 3D porosity volume, reservoir characterization .

Supervised artificial neural network-based method for conversion of solar radiation data (case study: Algeria)

NASA Astrophysics Data System (ADS)

Laidi, Maamar; Hanini, Salah; Rezrazi, Ahmed; Yaiche, Mohamed Redha; El Hadj, Abdallah Abdallah; Chellali, Farouk

2017-04-01

In this study, a backpropagation artificial neural network (BP-ANN) model is used as an alternative approach to predict solar radiation on tilted surfaces (SRT) using a number of variables involved in physical process. These variables are namely the latitude of the site, mean temperature and relative humidity, Linke turbidity factor and Angstrom coefficient, extraterrestrial solar radiation, solar radiation data measured on horizontal surfaces (SRH), and solar zenith angle. Experimental solar radiation data from 13 stations spread all over Algeria around the year (2004) were used for training/validation and testing the artificial neural networks (ANNs), and one station was used to make the interpolation of the designed ANN. The ANN model was trained, validated, and tested using 60, 20, and 20 % of all data, respectively. The configuration 8-35-1 (8 inputs, 35 hidden, and 1 output neurons) presented an excellent agreement between the prediction and the experimental data during the test stage with determination coefficient of 0.99 and root meat squared error of 5.75 Wh/m2, considering a three-layer feedforward backpropagation neural network with Levenberg-Marquardt training algorithm, a hyperbolic tangent sigmoid and linear transfer function at the hidden and the output layer, respectively. This novel model could be used by researchers or scientists to design high-efficiency solar devices that are usually tilted at an optimum angle to increase the solar incident on the surface.

A Deep Learning Approach to LIBS Spectroscopy for Planetary Applications

NASA Astrophysics Data System (ADS)

Mullen, T. H.; Parente, M.; Gemp, I.; Dyar, M. D.

2017-12-01

The ChemCam instrument on the Curiousity rover has collected >440,000 laser-induced breakdown spectra (LIBS) from 1500 different geological targets since 2012. The team is using a pipeline of preprocessing and partial least squares techniques to predict compositions of surface materials [1]. Unfortunately, such multivariate techniques are plagued by hard-to-meet assumptions involving constant hyperparameter tuning to specific elements and the amount of training data available; if the whole distribution of data is not seen, the method will overfit to the training data and generalizability will suffer. The rover only has 10 calibration targets on-board that represent a small subset of the geochemical samples the rover is expected to investigate. Deep neural networks have been used to bypass these issues in other fields. Semi-supervised techniques allow researchers to utilized small labeled datasets and vast amounts of unlabeled data. One example is the variational autoencoder model, a semi-supervised generative model in the form of a deep neural network. The autoencoder assumes that LIBS spectra are generated from a distribution conditioned on the elemental compositions in the sample and some nuisance. The system is broken into two models: one that predicts elemental composition from the spectra and one that generates spectra from compositions that may or may not be seen in the training set. The synthesized spectra show strong agreement with geochemical conventions to express specific compositions. The predictions of composition show improved generalizability to PLS. Deep neural networks have also been used to transfer knowledge from one dataset to another to solve unlabeled data problems. Given that vast amounts of laboratry LIBS spectra have been obtained in the past few years, it is now feasible train a deep net to predict elemental composition from lab spectra. Transfer learning (manifold alignment or calibration transfer) [2] is then used to fine-tune the model from terrestrial lab data to Martian field data. Neural networks and generative models provide the flexibility need for elemental composition prediction and unseen spectra synthesis. [1] Clegg S. et al. (2016) Spectrochim. Acta B, 129, 64-85. [2] Boucher T. et al. (2017) J. Chemom., 31, e2877.

Conduction Delay Learning Model for Unsupervised and Supervised Classification of Spatio-Temporal Spike Patterns

PubMed Central

Matsubara, Takashi

2017-01-01

Precise spike timing is considered to play a fundamental role in communications and signal processing in biological neural networks. Understanding the mechanism of spike timing adjustment would deepen our understanding of biological systems and enable advanced engineering applications such as efficient computational architectures. However, the biological mechanisms that adjust and maintain spike timing remain unclear. Existing algorithms adopt a supervised approach, which adjusts the axonal conduction delay and synaptic efficacy until the spike timings approximate the desired timings. This study proposes a spike timing-dependent learning model that adjusts the axonal conduction delay and synaptic efficacy in both unsupervised and supervised manners. The proposed learning algorithm approximates the Expectation-Maximization algorithm, and classifies the input data encoded into spatio-temporal spike patterns. Even in the supervised classification, the algorithm requires no external spikes indicating the desired spike timings unlike existing algorithms. Furthermore, because the algorithm is consistent with biological models and hypotheses found in existing biological studies, it could capture the mechanism underlying biological delay learning. PMID:29209191

Conduction Delay Learning Model for Unsupervised and Supervised Classification of Spatio-Temporal Spike Patterns.

PubMed

Matsubara, Takashi

2017-01-01

Precise spike timing is considered to play a fundamental role in communications and signal processing in biological neural networks. Understanding the mechanism of spike timing adjustment would deepen our understanding of biological systems and enable advanced engineering applications such as efficient computational architectures. However, the biological mechanisms that adjust and maintain spike timing remain unclear. Existing algorithms adopt a supervised approach, which adjusts the axonal conduction delay and synaptic efficacy until the spike timings approximate the desired timings. This study proposes a spike timing-dependent learning model that adjusts the axonal conduction delay and synaptic efficacy in both unsupervised and supervised manners. The proposed learning algorithm approximates the Expectation-Maximization algorithm, and classifies the input data encoded into spatio-temporal spike patterns. Even in the supervised classification, the algorithm requires no external spikes indicating the desired spike timings unlike existing algorithms. Furthermore, because the algorithm is consistent with biological models and hypotheses found in existing biological studies, it could capture the mechanism underlying biological delay learning.

Spiking Neural Networks Based on OxRAM Synapses for Real-Time Unsupervised Spike Sorting.

PubMed

Werner, Thilo; Vianello, Elisa; Bichler, Olivier; Garbin, Daniele; Cattaert, Daniel; Yvert, Blaise; De Salvo, Barbara; Perniola, Luca

2016-01-01

In this paper, we present an alternative approach to perform spike sorting of complex brain signals based on spiking neural networks (SNN). The proposed architecture is suitable for hardware implementation by using resistive random access memory (RRAM) technology for the implementation of synapses whose low latency (<1μs) enables real-time spike sorting. This offers promising advantages to conventional spike sorting techniques for brain-computer interfaces (BCI) and neural prosthesis applications. Moreover, the ultra-low power consumption of the RRAM synapses of the spiking neural network (nW range) may enable the design of autonomous implantable devices for rehabilitation purposes. We demonstrate an original methodology to use Oxide based RRAM (OxRAM) as easy to program and low energy (<75 pJ) synapses. Synaptic weights are modulated through the application of an online learning strategy inspired by biological Spike Timing Dependent Plasticity. Real spiking data have been recorded both intra- and extracellularly from an in-vitro preparation of the Crayfish sensory-motor system and used for validation of the proposed OxRAM based SNN. This artificial SNN is able to identify, learn, recognize and distinguish between different spike shapes in the input signal with a recognition rate about 90% without any supervision.

Adaptive template generation for amyloid PET using a deep learning approach.

PubMed

Kang, Seung Kwan; Seo, Seongho; Shin, Seong A; Byun, Min Soo; Lee, Dong Young; Kim, Yu Kyeong; Lee, Dong Soo; Lee, Jae Sung

2018-05-11

Accurate spatial normalization (SN) of amyloid positron emission tomography (PET) images for Alzheimer's disease assessment without coregistered anatomical magnetic resonance imaging (MRI) of the same individual is technically challenging. In this study, we applied deep neural networks to generate individually adaptive PET templates for robust and accurate SN of amyloid PET without using matched 3D MR images. Using 681 pairs of simultaneously acquired 11 C-PIB PET and T1-weighted 3D MRI scans of AD, MCI, and cognitively normal subjects, we trained and tested two deep neural networks [convolutional auto-encoder (CAE) and generative adversarial network (GAN)] that produce adaptive best PET templates. More specifically, the networks were trained using 685,100 pieces of augmented data generated by rotating 527 randomly selected datasets and validated using 154 datasets. The input to the supervised neural networks was the 3D PET volume in native space and the label was the spatially normalized 3D PET image using the transformation parameters obtained from MRI-based SN. The proposed deep learning approach significantly enhanced the quantitative accuracy of MRI-less amyloid PET assessment by reducing the SN error observed when an average amyloid PET template is used. Given an input image, the trained deep neural networks rapidly provide individually adaptive 3D PET templates without any discontinuity between the slices (in 0.02 s). As the proposed method does not require 3D MRI for the SN of PET images, it has great potential for use in routine analysis of amyloid PET images in clinical practice and research. © 2018 Wiley Periodicals, Inc.

FDI and Accommodation Using NN Based Techniques

NASA Astrophysics Data System (ADS)

Garcia, Ramon Ferreiro; de Miguel Catoira, Alberto; Sanz, Beatriz Ferreiro

Massive application of dynamic backpropagation neural networks is used on closed loop control FDI (fault detection and isolation) tasks. The process dynamics is mapped by means of a trained backpropagation NN to be applied on residual generation. Process supervision is then applied to discriminate faults on process sensors, and process plant parameters. A rule based expert system is used to implement the decision making task and the corresponding solution in terms of faults accommodation and/or reconfiguration. Results show an efficient and robust FDI system which could be used as the core of an SCADA or alternatively as a complement supervision tool operating in parallel with the SCADA when applied on a heat exchanger.

Learning to read aloud: A neural network approach using sparse distributed memory

NASA Technical Reports Server (NTRS)

Joglekar, Umesh Dwarkanath

1989-01-01

An attempt to solve a problem of text-to-phoneme mapping is described which does not appear amenable to solution by use of standard algorithmic procedures. Experiments based on a model of distributed processing are also described. This model (sparse distributed memory (SDM)) can be used in an iterative supervised learning mode to solve the problem. Additional improvements aimed at obtaining better performance are suggested.

Centered Kernel Alignment Enhancing Neural Network Pretraining for MRI-Based Dementia Diagnosis

PubMed Central

Cárdenas-Peña, David; Collazos-Huertas, Diego; Castellanos-Dominguez, German

2016-01-01

Dementia is a growing problem that affects elderly people worldwide. More accurate evaluation of dementia diagnosis can help during the medical examination. Several methods for computer-aided dementia diagnosis have been proposed using resonance imaging scans to discriminate between patients with Alzheimer's disease (AD) or mild cognitive impairment (MCI) and healthy controls (NC). Nonetheless, the computer-aided diagnosis is especially challenging because of the heterogeneous and intermediate nature of MCI. We address the automated dementia diagnosis by introducing a novel supervised pretraining approach that takes advantage of the artificial neural network (ANN) for complex classification tasks. The proposal initializes an ANN based on linear projections to achieve more discriminating spaces. Such projections are estimated by maximizing the centered kernel alignment criterion that assesses the affinity between the resonance imaging data kernel matrix and the label target matrix. As a result, the performed linear embedding allows accounting for features that contribute the most to the MCI class discrimination. We compare the supervised pretraining approach to two unsupervised initialization methods (autoencoders and Principal Component Analysis) and against the best four performing classification methods of the 2014 CADDementia challenge. As a result, our proposal outperforms all the baselines (7% of classification accuracy and area under the receiver-operating-characteristic curve) at the time it reduces the class biasing. PMID:27148392

A Universal 3D Voxel Descriptor for Solid-State Material Informatics with Deep Convolutional Neural Networks.

PubMed

Kajita, Seiji; Ohba, Nobuko; Jinnouchi, Ryosuke; Asahi, Ryoji

2017-12-05

Material informatics (MI) is a promising approach to liberate us from the time-consuming Edisonian (trial and error) process for material discoveries, driven by machine-learning algorithms. Several descriptors, which are encoded material features to feed computers, were proposed in the last few decades. Especially to solid systems, however, their insufficient representations of three dimensionality of field quantities such as electron distributions and local potentials have critically hindered broad and practical successes of the solid-state MI. We develop a simple, generic 3D voxel descriptor that compacts any field quantities, in such a suitable way to implement convolutional neural networks (CNNs). We examine the 3D voxel descriptor encoded from the electron distribution by a regression test with 680 oxides data. The present scheme outperforms other existing descriptors in the prediction of Hartree energies that are significantly relevant to the long-wavelength distribution of the valence electrons. The results indicate that this scheme can forecast any functionals of field quantities just by learning sufficient amount of data, if there is an explicit correlation between the target properties and field quantities. This 3D descriptor opens a way to import prominent CNNs-based algorithms of supervised, semi-supervised and reinforcement learnings into the solid-state MI.

Training Feedforward Neural Networks Using Symbiotic Organisms Search Algorithm

PubMed Central

Wu, Haizhou; Luo, Qifang

2016-01-01

Symbiotic organisms search (SOS) is a new robust and powerful metaheuristic algorithm, which stimulates the symbiotic interaction strategies adopted by organisms to survive and propagate in the ecosystem. In the supervised learning area, it is a challenging task to present a satisfactory and efficient training algorithm for feedforward neural networks (FNNs). In this paper, SOS is employed as a new method for training FNNs. To investigate the performance of the aforementioned method, eight different datasets selected from the UCI machine learning repository are employed for experiment and the results are compared among seven metaheuristic algorithms. The results show that SOS performs better than other algorithms for training FNNs in terms of converging speed. It is also proven that an FNN trained by the method of SOS has better accuracy than most algorithms compared. PMID:28105044

APPLICATION OF NEURAL NETWORK ALGORITHMS FOR BPM LINEARIZATION

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

Musson, John C.; Seaton, Chad; Spata, Mike F.

2012-11-01

Stripline BPM sensors contain inherent non-linearities, as a result of field distortions from the pickup elements. Many methods have been devised to facilitate corrections, often employing polynomial fitting. The cost of computation makes real-time correction difficult, particulalry when integer math is utilized. The application of neural-network technology, particularly the multi-layer perceptron algorithm, is proposed as an efficient alternative for electrode linearization. A process of supervised learning is initially used to determine the weighting coefficients, which are subsequently applied to the incoming electrode data. A non-linear layer, known as an activation layer, is responsible for the removal of saturation effects. Implementationmore » of a perceptron in an FPGA-based software-defined radio (SDR) is presented, along with performance comparisons. In addition, efficient calculation of the sigmoidal activation function via the CORDIC algorithm is presented.« less

Hyperspectral Image Classification With Markov Random Fields and a Convolutional Neural Network

NASA Astrophysics Data System (ADS)

Cao, Xiangyong; Zhou, Feng; Xu, Lin; Meng, Deyu; Xu, Zongben; Paisley, John

2018-05-01

This paper presents a new supervised classification algorithm for remotely sensed hyperspectral image (HSI) which integrates spectral and spatial information in a unified Bayesian framework. First, we formulate the HSI classification problem from a Bayesian perspective. Then, we adopt a convolutional neural network (CNN) to learn the posterior class distributions using a patch-wise training strategy to better use the spatial information. Next, spatial information is further considered by placing a spatial smoothness prior on the labels. Finally, we iteratively update the CNN parameters using stochastic gradient decent (SGD) and update the class labels of all pixel vectors using an alpha-expansion min-cut-based algorithm. Compared with other state-of-the-art methods, the proposed classification method achieves better performance on one synthetic dataset and two benchmark HSI datasets in a number of experimental settings.

Improving subjective pattern recognition in chemical senses through reduction of nonlinear effects in evaluation of sparse data

NASA Astrophysics Data System (ADS)

Assadi, Amir H.; Rasouli, Firooz; Wrenn, Susan E.; Subbiah, M.

2002-11-01

Artificial neural network models are typically useful in pattern recognition and extraction of important features in large data sets. These models are implemented in a wide variety of contexts and with diverse type of input-output data. The underlying mathematics of supervised training of neural networks is ultimately tied to the ability to approximate the nonlinearities that are inherent in network"s generalization ability. The quality and availability of sufficient data points for training and validation play a key role in the generalization ability of the network. A potential domain of applications of neural networks is in analysis of subjective data, such as in consumer science, affective neuroscience and perception of chemical senses. In applications of ANN to subjective data, it is common to rely on knowledge of the science and context for data acquisition, for instance as a priori probabilities in the Bayesian framework. In this paper, we discuss the circumstances that create challenges for success of neural network models for subjective data analysis, such as sparseness of data and cost of acquisition of additional samples. In particular, in the case of affect and perception of chemical senses, we suggest that inherent ambiguity of subjective responses could be offset by a combination of human-machine expert. We propose a method of pre- and post-processing for blind analysis of data that that relies on heuristics from human performance in interpretation of data. In particular, we offer an information-theoretic smoothing (ITS) algorithm that optimizes that geometric visualization of multi-dimensional data and improves human interpretation of the input-output view of neural network implementations. The pre- and post-processing algorithms and ITS are unsupervised. Finally, we discuss the details of an example of blind data analysis from actual taste-smell subjective data, and demonstrate the usefulness of PCA in reduction of dimensionality, as well as ITS.

Identifing Atmospheric Pollutant Sources Using Artificial Neural Networks

NASA Astrophysics Data System (ADS)

Paes, F. F.; Campos, H. F.; Luz, E. P.; Carvalho, A. R.

2008-05-01

The estimation of the area source pollutant strength is a relevant issue for atmospheric environment. This characterizes an inverse problem in the atmospheric pollution dispersion. In the inverse analysis, an area source domain is considered, where the strength of such area source term is assumed unknown. The inverse problem is solved by using a supervised artificial neural network: multi-layer perceptron. The conection weights of the neural network are computed from delta rule - learning process. The neural network inversion is compared with results from standard inverse analysis (regularized inverse solution). In the regularization method, the inverse problem is formulated as a non-linear optimization approach, whose the objective function is given by the square difference between the measured pollutant concentration and the mathematical models, associated with a regularization operator. In our numerical experiments, the forward problem is addressed by a source-receptor scheme, where a regressive Lagrangian model is applied to compute the transition matrix. The second order maximum entropy regularization is used, and the regularization parameter is calculated by the L-curve technique. The objective function is minimized employing a deterministic scheme (a quasi-Newton algorithm) [1] and a stochastic technique (PSO: particle swarm optimization) [2]. The inverse problem methodology is tested with synthetic observational data, from six measurement points in the physical domain. The best inverse solutions were obtained with neural networks. References: [1] D. R. Roberti, D. Anfossi, H. F. Campos Velho, G. A. Degrazia (2005): Estimating Emission Rate and Pollutant Source Location, Ciencia e Natura, p. 131-134. [2] E.F.P. da Luz, H.F. de Campos Velho, J.C. Becceneri, D.R. Roberti (2007): Estimating Atmospheric Area Source Strength Through Particle Swarm Optimization. Inverse Problems, Desing and Optimization Symposium IPDO-2007, April 16-18, Miami (FL), USA, vol 1, p. 354-359.

Application of the artificial neural network in quantitative structure-gradient elution retention relationship of phenylthiocarbamyl amino acids derivatives.

PubMed

Tham, S Y; Agatonovic-Kustrin, S

2002-05-15

Quantitative structure-retention relationship(QSRR) method was used to model reversed-phase high-performance liquid chromatography (RP-HPLC) separation of 18 selected amino acids. Retention data for phenylthiocarbamyl (PTC) amino acids derivatives were obtained using gradient elution on ODS column with mobile phase of varying acetonitrile, acetate buffer and containing 0.5 ml/l of triethylamine (TEA). Molecular structure of each amino acid was encoded with 36 calculated molecular descriptors. The correlation between the molecular descriptors and the retention time of the compounds in the calibration set was established using the genetic neural network method. A genetic algorithm (GA) was used to select important molecular descriptors and supervised artificial neural network (ANN) was used to correlate mobile phase composition and selected descriptors with the experimentally derived retention times. Retention time values were used as the network's output and calculated molecular descriptors and mobile phase composition as the inputs. The best model with five input descriptors was chosen, and the significance of the selected descriptors for amino acid separation was examined. Results confirmed the dominant role of the organic modifier in such chromatographic systems in addition to lipophilicity (log P) and molecular size and shape (topological indices) of investigated solutes.

One-dimensional inversion of geo-electrical resistivity sounding data using artificial neural networks—a case study

NASA Astrophysics Data System (ADS)

Singh, U. K.; Tiwari, R. K.; Singh, S. B.

2005-02-01

This paper deals with the application of artificial neural networks (ANN) technique for the study of a case history using 1-D inversion of vertical electrical resistivity sounding (VES) data from the Puga valley, Kashmir, India. The study area is important for its rich geothermal resources as well as from the tectonic point of view as it is located near the collision boundary of the Indo-Asian crustal plates. In order to understand the resistivity structure and layer thicknesses, we used here three-layer feedforward neural networks to model and predict measured VES data. Three algorithms, e.g. back-propagation (BP), adaptive back-propagation (ABP) and Levenberg-Marquardt algorithm (LMA) were applied to the synthetic as well as real VES field data and efficiency of supervised training network are compared. Analyses suggest that LMA is computationally faster and give results, which are comparatively more accurate and consistent than BP and ABP. The results obtained using the ANN inversions are remarkably correlated with the available borehole litho-logs. The feasibility study suggests that ANN methods offer an excellent complementary tool for the direct detection of layered resistivity structure.

CNN universal machine as classificaton platform: an art-like clustering algorithm.

PubMed

Bálya, David

2003-12-01

Fast and robust classification of feature vectors is a crucial task in a number of real-time systems. A cellular neural/nonlinear network universal machine (CNN-UM) can be very efficient as a feature detector. The next step is to post-process the results for object recognition. This paper shows how a robust classification scheme based on adaptive resonance theory (ART) can be mapped to the CNN-UM. Moreover, this mapping is general enough to include different types of feed-forward neural networks. The designed analogic CNN algorithm is capable of classifying the extracted feature vectors keeping the advantages of the ART networks, such as robust, plastic and fault-tolerant behaviors. An analogic algorithm is presented for unsupervised classification with tunable sensitivity and automatic new class creation. The algorithm is extended for supervised classification. The presented binary feature vector classification is implemented on the existing standard CNN-UM chips for fast classification. The experimental evaluation shows promising performance after 100% accuracy on the training set.

Seismic waveform classification using deep learning

NASA Astrophysics Data System (ADS)

Kong, Q.; Allen, R. M.

2017-12-01

MyShake is a global smartphone seismic network that harnesses the power of crowdsourcing. It has an Artificial Neural Network (ANN) algorithm running on the phone to distinguish earthquake motion from human activities recorded by the accelerometer on board. Once the ANN detects earthquake-like motion, it sends a 5-min chunk of acceleration data back to the server for further analysis. The time-series data collected contains both earthquake data and human activity data that the ANN confused. In this presentation, we will show the Convolutional Neural Network (CNN) we built under the umbrella of supervised learning to find out the earthquake waveform. The waveforms of the recorded motion could treat easily as images, and by taking the advantage of the power of CNN processing the images, we achieved very high successful rate to select the earthquake waveforms out. Since there are many non-earthquake waveforms than the earthquake waveforms, we also built an anomaly detection algorithm using the CNN. Both these two methods can be easily extended to other waveform classification problems.

Action-Driven Visual Object Tracking With Deep Reinforcement Learning.

PubMed

Yun, Sangdoo; Choi, Jongwon; Yoo, Youngjoon; Yun, Kimin; Choi, Jin Young

2018-06-01

In this paper, we propose an efficient visual tracker, which directly captures a bounding box containing the target object in a video by means of sequential actions learned using deep neural networks. The proposed deep neural network to control tracking actions is pretrained using various training video sequences and fine-tuned during actual tracking for online adaptation to a change of target and background. The pretraining is done by utilizing deep reinforcement learning (RL) as well as supervised learning. The use of RL enables even partially labeled data to be successfully utilized for semisupervised learning. Through the evaluation of the object tracking benchmark data set, the proposed tracker is validated to achieve a competitive performance at three times the speed of existing deep network-based trackers. The fast version of the proposed method, which operates in real time on graphics processing unit, outperforms the state-of-the-art real-time trackers with an accuracy improvement of more than 8%.

Distributed Learning, Recognition, and Prediction by ART and ARTMAP Neural Networks.

PubMed

Carpenter, Gail A.

1997-11-01

A class of adaptive resonance theory (ART) models for learning, recognition, and prediction with arbitrarily distributed code representations is introduced. Distributed ART neural networks combine the stable fast learning capabilities of winner-take-all ART systems with the noise tolerance and code compression capabilities of multilayer perceptrons. With a winner-take-all code, the unsupervised model dART reduces to fuzzy ART and the supervised model dARTMAP reduces to fuzzy ARTMAP. With a distributed code, these networks automatically apportion learned changes according to the degree of activation of each coding node, which permits fast as well as slow learning without catastrophic forgetting. Distributed ART models replace the traditional neural network path weight with a dynamic weight equal to the rectified difference between coding node activation and an adaptive threshold. Thresholds increase monotonically during learning according to a principle of atrophy due to disuse. However, monotonic change at the synaptic level manifests itself as bidirectional change at the dynamic level, where the result of adaptation resembles long-term potentiation (LTP) for single-pulse or low frequency test inputs but can resemble long-term depression (LTD) for higher frequency test inputs. This paradoxical behavior is traced to dual computational properties of phasic and tonic coding signal components. A parallel distributed match-reset-search process also helps stabilize memory. Without the match-reset-search system, dART becomes a type of distributed competitive learning network.

Multilabel user classification using the community structure of online networks

PubMed Central

Papadopoulos, Symeon; Kompatsiaris, Yiannis

2017-01-01

We study the problem of semi-supervised, multi-label user classification of networked data in the online social platform setting. We propose a framework that combines unsupervised community extraction and supervised, community-based feature weighting before training a classifier. We introduce Approximate Regularized Commute-Time Embedding (ARCTE), an algorithm that projects the users of a social graph onto a latent space, but instead of packing the global structure into a matrix of predefined rank, as many spectral and neural representation learning methods do, it extracts local communities for all users in the graph in order to learn a sparse embedding. To this end, we employ an improvement of personalized PageRank algorithms for searching locally in each user’s graph structure. Then, we perform supervised community feature weighting in order to boost the importance of highly predictive communities. We assess our method performance on the problem of user classification by performing an extensive comparative study among various recent methods based on graph embeddings. The comparison shows that ARCTE significantly outperforms the competition in almost all cases, achieving up to 35% relative improvement compared to the second best competing method in terms of F1-score. PMID:28278242

Multilabel user classification using the community structure of online networks.

PubMed

Rizos, Georgios; Papadopoulos, Symeon; Kompatsiaris, Yiannis

2017-01-01

We study the problem of semi-supervised, multi-label user classification of networked data in the online social platform setting. We propose a framework that combines unsupervised community extraction and supervised, community-based feature weighting before training a classifier. We introduce Approximate Regularized Commute-Time Embedding (ARCTE), an algorithm that projects the users of a social graph onto a latent space, but instead of packing the global structure into a matrix of predefined rank, as many spectral and neural representation learning methods do, it extracts local communities for all users in the graph in order to learn a sparse embedding. To this end, we employ an improvement of personalized PageRank algorithms for searching locally in each user's graph structure. Then, we perform supervised community feature weighting in order to boost the importance of highly predictive communities. We assess our method performance on the problem of user classification by performing an extensive comparative study among various recent methods based on graph embeddings. The comparison shows that ARCTE significantly outperforms the competition in almost all cases, achieving up to 35% relative improvement compared to the second best competing method in terms of F1-score.

The structure-AChE inhibitory activity relationships study in a series of pyridazine analogues.

PubMed

Saracoglu, M; Kandemirli, F

2009-07-01

The structure-activity relationships (SAR) are investigated by means of the Electronic-Topological Method (ETM) followed by the Neural Networks application (ETM-NN) for a class of anti-cholinesterase inhibitors (AChE, 53 molecules) being pyridazine derivatives. AChE activities of the series were measured in IC(50) units, and relative to the activity levels, the series was partitioned into classes of active and inactive compounds. Based on pharmacophores and antipharmacophores calculated by the ETM-software as sub-matrices containing important spatial and electronic characteristics, a system for the activity prognostication is developed. Input data for the ETM were taken as the results of conformational and quantum-mechanics calculations. To predict the activity, we used one of the most well known neural networks, namely, the feed-forward neural networks (FFNNs) trained with the back propagation algorithm. The supervised learning was performed using a variant of FFNN known as the Associative Neural Networks (ASNN). The result of the testing revealed that the high ETM's ability of predicting both activity and inactivity of potential AChE inhibitors. Analysis of HOMOs for the compounds containing Ph1 and APh1 has shown that atoms with the highest values of the atomic orbital coefficients are mainly those atoms that enter into the pharmacophores. Thus, the set of pharmacophores and antipharmacophores found as the result of this study forms a basis for a system of the anti-cholinesterase activity prediction.

Sediment classification using neural networks: An example from the site-U1344A of IODP Expedition 323 in the Bering Sea

NASA Astrophysics Data System (ADS)

Ojha, Maheswar; Maiti, Saumen

2016-03-01

A novel approach based on the concept of Bayesian neural network (BNN) has been implemented for classifying sediment boundaries using downhole log data obtained during Integrated Ocean Drilling Program (IODP) Expedition 323 in the Bering Sea slope region. The Bayesian framework in conjunction with Markov Chain Monte Carlo (MCMC)/hybrid Monte Carlo (HMC) learning paradigm has been applied to constrain the lithology boundaries using density, density porosity, gamma ray, sonic P-wave velocity and electrical resistivity at the Hole U1344A. We have demonstrated the effectiveness of our supervised classification methodology by comparing our findings with a conventional neural network and a Bayesian neural network optimized by scaled conjugate gradient method (SCG), and tested the robustness of the algorithm in the presence of red noise in the data. The Bayesian results based on the HMC algorithm (BNN.HMC) resolve detailed finer structures at certain depths in addition to main lithology such as silty clay, diatom clayey silt and sandy silt. Our method also recovers the lithology information from a depth ranging between 615 and 655 m Wireline log Matched depth below Sea Floor of no core recovery zone. Our analyses demonstrate that the BNN based approach renders robust means for the classification of complex lithology successions at the Hole U1344A, which could be very useful for other studies and understanding the oceanic crustal inhomogeneity and structural discontinuities.

Efficient Training of Supervised Spiking Neural Network via Accurate Synaptic-Efficiency Adjustment Method.

PubMed

Xie, Xiurui; Qu, Hong; Yi, Zhang; Kurths, Jurgen

2017-06-01

The spiking neural network (SNN) is the third generation of neural networks and performs remarkably well in cognitive tasks, such as pattern recognition. The temporal neural encode mechanism found in biological hippocampus enables SNN to possess more powerful computation capability than networks with other encoding schemes. However, this temporal encoding approach requires neurons to process information serially on time, which reduces learning efficiency significantly. To keep the powerful computation capability of the temporal encoding mechanism and to overcome its low efficiency in the training of SNNs, a new training algorithm, the accurate synaptic-efficiency adjustment method is proposed in this paper. Inspired by the selective attention mechanism of the primate visual system, our algorithm selects only the target spike time as attention areas, and ignores voltage states of the untarget ones, resulting in a significant reduction of training time. Besides, our algorithm employs a cost function based on the voltage difference between the potential of the output neuron and the firing threshold of the SNN, instead of the traditional precise firing time distance. A normalized spike-timing-dependent-plasticity learning window is applied to assigning this error to different synapses for instructing their training. Comprehensive simulations are conducted to investigate the learning properties of our algorithm, with input neurons emitting both single spike and multiple spikes. Simulation results indicate that our algorithm possesses higher learning performance than the existing other methods and achieves the state-of-the-art efficiency in the training of SNN.

Study on recognition algorithm for paper currency numbers based on neural network

NASA Astrophysics Data System (ADS)

Li, Xiuyan; Liu, Tiegen; Li, Yuanyao; Zhang, Zhongchuan; Deng, Shichao

2008-12-01

Based on the unique characteristic, the paper currency numbers can be put into record and the automatic identification equipment for paper currency numbers is supplied to currency circulation market in order to provide convenience for financial sectors to trace the fiduciary circulation socially and provide effective supervision on paper currency. Simultaneously it is favorable for identifying forged notes, blacklisting the forged notes numbers and solving the major social problems, such as armor cash carrier robbery, money laundering. For the purpose of recognizing the paper currency numbers, a recognition algorithm based on neural network is presented in the paper. Number lines in original paper currency images can be draw out through image processing, such as image de-noising, skew correction, segmentation, and image normalization. According to the different characteristics between digits and letters in serial number, two kinds of classifiers are designed. With the characteristics of associative memory, optimization-compute and rapid convergence, the Discrete Hopfield Neural Network (DHNN) is utilized to recognize the letters; with the characteristics of simple structure, quick learning and global optimum, the Radial-Basis Function Neural Network (RBFNN) is adopted to identify the digits. Then the final recognition results are obtained by combining the two kinds of recognition results in regular sequence. Through the simulation tests, it is confirmed by simulation results that the recognition algorithm of combination of two kinds of recognition methods has such advantages as high recognition rate and faster recognition simultaneously, which is worthy of broad application prospect.

Deep Learning Automates the Quantitative Analysis of Individual Cells in Live-Cell Imaging Experiments.

PubMed

Van Valen, David A; Kudo, Takamasa; Lane, Keara M; Macklin, Derek N; Quach, Nicolas T; DeFelice, Mialy M; Maayan, Inbal; Tanouchi, Yu; Ashley, Euan A; Covert, Markus W

2016-11-01

Live-cell imaging has opened an exciting window into the role cellular heterogeneity plays in dynamic, living systems. A major critical challenge for this class of experiments is the problem of image segmentation, or determining which parts of a microscope image correspond to which individual cells. Current approaches require many hours of manual curation and depend on approaches that are difficult to share between labs. They are also unable to robustly segment the cytoplasms of mammalian cells. Here, we show that deep convolutional neural networks, a supervised machine learning method, can solve this challenge for multiple cell types across the domains of life. We demonstrate that this approach can robustly segment fluorescent images of cell nuclei as well as phase images of the cytoplasms of individual bacterial and mammalian cells from phase contrast images without the need for a fluorescent cytoplasmic marker. These networks also enable the simultaneous segmentation and identification of different mammalian cell types grown in co-culture. A quantitative comparison with prior methods demonstrates that convolutional neural networks have improved accuracy and lead to a significant reduction in curation time. We relay our experience in designing and optimizing deep convolutional neural networks for this task and outline several design rules that we found led to robust performance. We conclude that deep convolutional neural networks are an accurate method that require less curation time, are generalizable to a multiplicity of cell types, from bacteria to mammalian cells, and expand live-cell imaging capabilities to include multi-cell type systems.

Deep Learning Automates the Quantitative Analysis of Individual Cells in Live-Cell Imaging Experiments

DOE PAGES

Van Valen, David A.; Kudo, Takamasa; Lane, Keara M.; ...

2016-11-04

Live-cell imaging has opened an exciting window into the role cellular heterogeneity plays in dynamic, living systems. A major critical challenge for this class of experiments is the problem of image segmentation, or determining which parts of a microscope image correspond to which individual cells. Current approaches require many hours of manual curation and depend on approaches that are difficult to share between labs. They are also unable to robustly segment the cytoplasms of mammalian cells. Here, we show that deep convolutional neural networks, a supervised machine learning method, can solve this challenge for multiple cell types across the domainsmore » of life. We demonstrate that this approach can robustly segment fluorescent images of cell nuclei as well as phase images of the cytoplasms of individual bacterial and mammalian cells from phase contrast images without the need for a fluorescent cytoplasmic marker. These networks also enable the simultaneous segmentation and identification of different mammalian cell types grown in co-culture. A quantitative comparison with prior methods demonstrates that convolutional neural networks have improved accuracy and lead to a significant reduction in curation time. We relay our experience in designing and optimizing deep convolutional neural networks for this task and outline several design rules that we found led to robust performance. We conclude that deep convolutional neural networks are an accurate method that require less curation time, are generalizable to a multiplicity of cell types, from bacteria to mammalian cells, and expand live-cell imaging capabilities to include multi-cell type systems.« less

Deep Learning Automates the Quantitative Analysis of Individual Cells in Live-Cell Imaging Experiments

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

Van Valen, David A.; Kudo, Takamasa; Lane, Keara M.

Live-cell imaging has opened an exciting window into the role cellular heterogeneity plays in dynamic, living systems. A major critical challenge for this class of experiments is the problem of image segmentation, or determining which parts of a microscope image correspond to which individual cells. Current approaches require many hours of manual curation and depend on approaches that are difficult to share between labs. They are also unable to robustly segment the cytoplasms of mammalian cells. Here, we show that deep convolutional neural networks, a supervised machine learning method, can solve this challenge for multiple cell types across the domainsmore » of life. We demonstrate that this approach can robustly segment fluorescent images of cell nuclei as well as phase images of the cytoplasms of individual bacterial and mammalian cells from phase contrast images without the need for a fluorescent cytoplasmic marker. These networks also enable the simultaneous segmentation and identification of different mammalian cell types grown in co-culture. A quantitative comparison with prior methods demonstrates that convolutional neural networks have improved accuracy and lead to a significant reduction in curation time. We relay our experience in designing and optimizing deep convolutional neural networks for this task and outline several design rules that we found led to robust performance. We conclude that deep convolutional neural networks are an accurate method that require less curation time, are generalizable to a multiplicity of cell types, from bacteria to mammalian cells, and expand live-cell imaging capabilities to include multi-cell type systems.« less

Deep Learning Automates the Quantitative Analysis of Individual Cells in Live-Cell Imaging Experiments

PubMed Central

Van Valen, David A.; Lane, Keara M.; Quach, Nicolas T.; Maayan, Inbal

2016-01-01

Live-cell imaging has opened an exciting window into the role cellular heterogeneity plays in dynamic, living systems. A major critical challenge for this class of experiments is the problem of image segmentation, or determining which parts of a microscope image correspond to which individual cells. Current approaches require many hours of manual curation and depend on approaches that are difficult to share between labs. They are also unable to robustly segment the cytoplasms of mammalian cells. Here, we show that deep convolutional neural networks, a supervised machine learning method, can solve this challenge for multiple cell types across the domains of life. We demonstrate that this approach can robustly segment fluorescent images of cell nuclei as well as phase images of the cytoplasms of individual bacterial and mammalian cells from phase contrast images without the need for a fluorescent cytoplasmic marker. These networks also enable the simultaneous segmentation and identification of different mammalian cell types grown in co-culture. A quantitative comparison with prior methods demonstrates that convolutional neural networks have improved accuracy and lead to a significant reduction in curation time. We relay our experience in designing and optimizing deep convolutional neural networks for this task and outline several design rules that we found led to robust performance. We conclude that deep convolutional neural networks are an accurate method that require less curation time, are generalizable to a multiplicity of cell types, from bacteria to mammalian cells, and expand live-cell imaging capabilities to include multi-cell type systems. PMID:27814364

Artificial Neural Networks as an Architectural Design Tool-Generating New Detail Forms Based On the Roman Corinthian Order Capital

NASA Astrophysics Data System (ADS)

Radziszewski, Kacper

2017-10-01

The following paper presents the results of the research in the field of the machine learning, investigating the scope of application of the artificial neural networks algorithms as a tool in architectural design. The computational experiment was held using the backward propagation of errors method of training the artificial neural network, which was trained based on the geometry of the details of the Roman Corinthian order capital. During the experiment, as an input training data set, five local geometry parameters combined has given the best results: Theta, Pi, Rho in spherical coordinate system based on the capital volume centroid, followed by Z value of the Cartesian coordinate system and a distance from vertical planes created based on the capital symmetry. Additionally during the experiment, artificial neural network hidden layers optimal count and structure was found, giving results of the error below 0.2% for the mentioned before input parameters. Once successfully trained artificial network, was able to mimic the details composition on any other geometry type given. Despite of calculating the transformed geometry locally and separately for each of the thousands of surface points, system could create visually attractive and diverse, complex patterns. Designed tool, based on the supervised learning method of machine learning, gives possibility of generating new architectural forms- free of the designer’s imagination bounds. Implementing the infinitely broad computational methods of machine learning, or Artificial Intelligence in general, not only could accelerate and simplify the design process, but give an opportunity to explore never seen before, unpredictable forms or everyday architectural practice solutions.

Supervised learning from human performance at the computationally hard problem of optimal traffic signal control on a network of junctions

PubMed Central

Box, Simon

2014-01-01

Optimal switching of traffic lights on a network of junctions is a computationally intractable problem. In this research, road traffic networks containing signallized junctions are simulated. A computer game interface is used to enable a human ‘player’ to control the traffic light settings on the junctions within the simulation. A supervised learning approach, based on simple neural network classifiers can be used to capture human player's strategies in the game and thus develop a human-trained machine control (HuTMaC) system that approaches human levels of performance. Experiments conducted within the simulation compare the performance of HuTMaC to two well-established traffic-responsive control systems that are widely deployed in the developed world and also to a temporal difference learning-based control method. In all experiments, HuTMaC outperforms the other control methods in terms of average delay and variance over delay. The conclusion is that these results add weight to the suggestion that HuTMaC may be a viable alternative, or supplemental method, to approximate optimization for some practical engineering control problems where the optimal strategy is computationally intractable. PMID:26064570

Supervised learning from human performance at the computationally hard problem of optimal traffic signal control on a network of junctions.

PubMed

Box, Simon

2014-12-01

Optimal switching of traffic lights on a network of junctions is a computationally intractable problem. In this research, road traffic networks containing signallized junctions are simulated. A computer game interface is used to enable a human 'player' to control the traffic light settings on the junctions within the simulation. A supervised learning approach, based on simple neural network classifiers can be used to capture human player's strategies in the game and thus develop a human-trained machine control (HuTMaC) system that approaches human levels of performance. Experiments conducted within the simulation compare the performance of HuTMaC to two well-established traffic-responsive control systems that are widely deployed in the developed world and also to a temporal difference learning-based control method. In all experiments, HuTMaC outperforms the other control methods in terms of average delay and variance over delay. The conclusion is that these results add weight to the suggestion that HuTMaC may be a viable alternative, or supplemental method, to approximate optimization for some practical engineering control problems where the optimal strategy is computationally intractable.

SKYNET: an efficient and robust neural network training tool for machine learning in astronomy

NASA Astrophysics Data System (ADS)

Graff, Philip; Feroz, Farhan; Hobson, Michael P.; Lasenby, Anthony

2014-06-01

We present the first public release of our generic neural network training algorithm, called SKYNET. This efficient and robust machine learning tool is able to train large and deep feed-forward neural networks, including autoencoders, for use in a wide range of supervised and unsupervised learning applications, such as regression, classification, density estimation, clustering and dimensionality reduction. SKYNET uses a `pre-training' method to obtain a set of network parameters that has empirically been shown to be close to a good solution, followed by further optimization using a regularized variant of Newton's method, where the level of regularization is determined and adjusted automatically; the latter uses second-order derivative information to improve convergence, but without the need to evaluate or store the full Hessian matrix, by using a fast approximate method to calculate Hessian-vector products. This combination of methods allows for the training of complicated networks that are difficult to optimize using standard backpropagation techniques. SKYNET employs convergence criteria that naturally prevent overfitting, and also includes a fast algorithm for estimating the accuracy of network outputs. The utility and flexibility of SKYNET are demonstrated by application to a number of toy problems, and to astronomical problems focusing on the recovery of structure from blurred and noisy images, the identification of gamma-ray bursters, and the compression and denoising of galaxy images. The SKYNET software, which is implemented in standard ANSI C and fully parallelized using MPI, is available at http://www.mrao.cam.ac.uk/software/skynet/.

The integration of elastic wave properties and machine learning for the distribution of petrophysical properties in reservoir modeling

NASA Astrophysics Data System (ADS)

Ratnam, T. C.; Ghosh, D. P.; Negash, B. M.

2018-05-01

Conventional reservoir modeling employs variograms to predict the spatial distribution of petrophysical properties. This study aims to improve property distribution by incorporating elastic wave properties. In this study, elastic wave properties obtained from seismic inversion are used as input for an artificial neural network to predict neutron porosity in between well locations. The method employed in this study is supervised learning based on available well logs. This method converts every seismic trace into a pseudo-well log, hence reducing the uncertainty between well locations. By incorporating the seismic response, the reliance on geostatistical methods such as variograms for the distribution of petrophysical properties is reduced drastically. The results of the artificial neural network show good correlation with the neutron porosity log which gives confidence for spatial prediction in areas where well logs are not available.

Digital Family History Data Mining with Neural Networks: A Pilot Study.

PubMed

Hoyt, Robert; Linnville, Steven; Thaler, Stephen; Moore, Jeffrey

2016-01-01

Following the passage of the Health Information Technology for Economic and Clinical Health (HITECH) Act of 2009, electronic health records were widely adopted by eligible physicians and hospitals in the United States. Stage 2 meaningful use menu objectives include a digital family history but no stipulation as to how that information should be used. A variety of data mining techniques now exist for these data, which include artificial neural networks (ANNs) for supervised or unsupervised machine learning. In this pilot study, we applied an ANN-based simulation to a previously reported digital family history to mine the database for trends. A graphical user interface was created to display the input of multiple conditions in the parents and output as the likelihood of diabetes, hypertension, and coronary artery disease in male and female offspring. The results of this pilot study show promise in using ANNs to data mine digital family histories for clinical and research purposes.

Condition monitoring of an electro-magnetic brake using an artificial neural network

NASA Astrophysics Data System (ADS)

Gofran, T.; Neugebauer, P.; Schramm, D.

2017-10-01

This paper presents a data-driven approach to Condition Monitoring of Electromagnetic brakes without use of additional sensors. For safe and efficient operation of electric motor a regular evaluation and replacement of the friction surface of the brake is required. One such evaluation method consists of direct or indirect sensing of the air-gap between pressure plate and magnet. A larger gap is generally indicative of worn surface(s). Traditionally this has been accomplished by the use of additional sensors - making existing systems complex, cost- sensitive and difficult to maintain. In this work a feed-forward Artificial Neural Network (ANN) is learned with the electrical data of the brake by supervised learning method to estimate the air-gap. The ANN model is optimized on the training set and validated using the test set. The experimental results of estimated air-gap with accuracy of over 95% demonstrate the validity of the proposed approach.

Enhancing deep convolutional neural network scheme for breast cancer diagnosis with unlabeled data.

PubMed

Sun, Wenqing; Tseng, Tzu-Liang Bill; Zhang, Jianying; Qian, Wei

2017-04-01

In this study we developed a graph based semi-supervised learning (SSL) scheme using deep convolutional neural network (CNN) for breast cancer diagnosis. CNN usually needs a large amount of labeled data for training and fine tuning the parameters, and our proposed scheme only requires a small portion of labeled data in training set. Four modules were included in the diagnosis system: data weighing, feature selection, dividing co-training data labeling, and CNN. 3158 region of interests (ROIs) with each containing a mass extracted from 1874 pairs of mammogram images were used for this study. Among them 100 ROIs were treated as labeled data while the rest were treated as unlabeled. The area under the curve (AUC) observed in our study was 0.8818, and the accuracy of CNN is 0.8243 using the mixed labeled and unlabeled data. Copyright © 2016. Published by Elsevier Ltd.

A new feature constituting approach to detection of vocal fold pathology

NASA Astrophysics Data System (ADS)

Hariharan, M.; Polat, Kemal; Yaacob, Sazali

2014-08-01

In the last two decades, non-invasive methods through acoustic analysis of voice signal have been proved to be excellent and reliable tool to diagnose vocal fold pathologies. This paper proposes a new feature vector based on the wavelet packet transform and singular value decomposition for the detection of vocal fold pathology. k-means clustering based feature weighting is proposed to increase the distinguishing performance of the proposed features. In this work, two databases Massachusetts Eye and Ear Infirmary (MEEI) voice disorders database and MAPACI speech pathology database are used. Four different supervised classifiers such as k-nearest neighbour (k-NN), least-square support vector machine, probabilistic neural network and general regression neural network are employed for testing the proposed features. The experimental results uncover that the proposed features give very promising classification accuracy of 100% for both MEEI database and MAPACI speech pathology database.

Neural Network Classification of Receiver Functions as a Step Towards Automatic Crustal Parameter Determination

NASA Astrophysics Data System (ADS)

Jemberie, A.; Dugda, M. T.; Reusch, D.; Nyblade, A.

2006-12-01

Neural networks are decision making mathematical/engineering tools, which if trained properly, can do jobs automatically (and objectively) that normally require particular expertise and/or tedious repetition. Here we explore two techniques from the field of artificial neural networks (ANNs) that seek to reduce the time requirements and increase the objectivity of quality control (QC) and Event Identification (EI) on seismic datasets. We explore to apply the multiplayer Feed Forward (FF) Artificial Neural Networks (ANN) and Self- Organizing Maps (SOM) in combination with Hk stacking of receiver functions in an attempt to test the extent of the usefulness of automatic classification of receiver functions for crustal parameter determination. Feed- forward ANNs (FFNNs) are a supervised classification tool while self-organizing maps (SOMs) are able to provide unsupervised classification of large, complex geophysical data sets into a fixed number of distinct generalized patterns or modes. Hk stacking is a methodology that is used to stack receiver functions based on the relative arrival times of P-to-S converted phase and next two reverberations to determine crustal thickness H and Vp-to-Vs ratio (k). We use receiver functions from teleseismic events recorded by the 2000- 2002 Ethiopia Broadband Seismic Experiment. Preliminary results of applying FFNN neural network and Hk stacking of receiver functions for automatic receiver functions classification as a step towards an effort of automatic crustal parameter determination look encouraging. After training a FFNN neural network, the network could classify the best receiver functions from bad ones with a success rate of about 75 to 95%. Applying H? stacking on the receiver functions classified by this FFNN as the best receiver functions, we could obtain crustal thickness and Vp/Vs ratio of 31±4 km and 1.75±0.05, respectively, for the crust beneath station ARBA in the Main Ethiopian Rift. To make comparison, we applied Hk stacking on the receiver functions which we ourselves classified as the best set and found that the crustal thickness and Vp/Vs ratio are 31±2 km and 1.75±0.02, respectively.

Error estimation of deformable image registration of pulmonary CT scans using convolutional neural networks.

PubMed

Eppenhof, Koen A J; Pluim, Josien P W

2018-04-01

Error estimation in nonlinear medical image registration is a nontrivial problem that is important for validation of registration methods. We propose a supervised method for estimation of registration errors in nonlinear registration of three-dimensional (3-D) images. The method is based on a 3-D convolutional neural network that learns to estimate registration errors from a pair of image patches. By applying the network to patches centered around every voxel, we construct registration error maps. The network is trained using a set of representative images that have been synthetically transformed to construct a set of image pairs with known deformations. The method is evaluated on deformable registrations of inhale-exhale pairs of thoracic CT scans. Using ground truth target registration errors on manually annotated landmarks, we evaluate the method's ability to estimate local registration errors. Estimation of full domain error maps is evaluated using a gold standard approach. The two evaluation approaches show that we can train the network to robustly estimate registration errors in a predetermined range, with subvoxel accuracy. We achieved a root-mean-square deviation of 0.51 mm from gold standard registration errors and of 0.66 mm from ground truth landmark registration errors.

Distributed multisensory integration in a recurrent network model through supervised learning

NASA Astrophysics Data System (ADS)

Wang, He; Wong, K. Y. Michael

Sensory integration between different modalities has been extensively studied. It is suggested that the brain integrates signals from different modalities in a Bayesian optimal way. However, how the Bayesian rule is implemented in a neural network remains under debate. In this work we propose a biologically plausible recurrent network model, which can perform Bayesian multisensory integration after trained by supervised learning. Our model is composed of two modules, each for one modality. We assume that each module is a recurrent network, whose activity represents the posterior distribution of each stimulus. The feedforward input on each module is the likelihood of each modality. Two modules are integrated through cross-links, which are feedforward connections from the other modality, and reciprocal connections, which are recurrent connections between different modules. By stochastic gradient descent, we successfully trained the feedforward and recurrent coupling matrices simultaneously, both of which resembles the Mexican-hat. We also find that there are more than one set of coupling matrices that can approximate the Bayesian theorem well. Specifically, reciprocal connections and cross-links will compensate each other if one of them is removed. Even though trained with two inputs, the network's performance with only one input is in good accordance with what is predicted by the Bayesian theorem.

Evaluation of machine learning algorithms for classification of primary biological aerosol using a new UV-LIF spectrometer

NASA Astrophysics Data System (ADS)

Ruske, Simon; Topping, David O.; Foot, Virginia E.; Kaye, Paul H.; Stanley, Warren R.; Crawford, Ian; Morse, Andrew P.; Gallagher, Martin W.

2017-03-01

Characterisation of bioaerosols has important implications within environment and public health sectors. Recent developments in ultraviolet light-induced fluorescence (UV-LIF) detectors such as the Wideband Integrated Bioaerosol Spectrometer (WIBS) and the newly introduced Multiparameter Bioaerosol Spectrometer (MBS) have allowed for the real-time collection of fluorescence, size and morphology measurements for the purpose of discriminating between bacteria, fungal spores and pollen.This new generation of instruments has enabled ever larger data sets to be compiled with the aim of studying more complex environments. In real world data sets, particularly those from an urban environment, the population may be dominated by non-biological fluorescent interferents, bringing into question the accuracy of measurements of quantities such as concentrations. It is therefore imperative that we validate the performance of different algorithms which can be used for the task of classification.For unsupervised learning we tested hierarchical agglomerative clustering with various different linkages. For supervised learning, 11 methods were tested, including decision trees, ensemble methods (random forests, gradient boosting and AdaBoost), two implementations for support vector machines (libsvm and liblinear) and Gaussian methods (Gaussian naïve Bayesian, quadratic and linear discriminant analysis, the k-nearest neighbours algorithm and artificial neural networks).The methods were applied to two different data sets produced using the new MBS, which provides multichannel UV-LIF fluorescence signatures for single airborne biological particles. The first data set contained mixed PSLs and the second contained a variety of laboratory-generated aerosol.Clustering in general performs slightly worse than the supervised learning methods, correctly classifying, at best, only 67. 6 and 91. 1 % for the two data sets respectively. For supervised learning the gradient boosting algorithm was found to be the most effective, on average correctly classifying 82. 8 and 98. 27 % of the testing data, respectively, across the two data sets.A possible alternative to gradient boosting is neural networks. We do however note that this method requires much more user input than the other methods, and we suggest that further research should be conducted using this method, especially using parallelised hardware such as the GPU, which would allow for larger networks to be trained, which could possibly yield better results.We also saw that some methods, such as clustering, failed to utilise the additional shape information provided by the instrument, whilst for others, such as the decision trees, ensemble methods and neural networks, improved performance could be attained with the inclusion of such information.

Visualization of suspicious lesions in breast MRI based on intelligent neural systems

NASA Astrophysics Data System (ADS)

Twellmann, Thorsten; Lange, Oliver; Nattkemper, Tim Wilhelm; Meyer-Bäse, Anke

2006-05-01

Intelligent medical systems based on supervised and unsupervised artificial neural networks are applied to the automatic visualization and classification of suspicious lesions in breast MRI. These systems represent an important component of future sophisticated computer-aided diagnosis systems and enable the extraction of spatial and temporal features of dynamic MRI data stemming from patients with confirmed lesion diagnosis. By taking into account the heterogenity of the cancerous tissue, these techniques reveal the malignant, benign and normal kinetic signals and and provide a regional subclassification of pathological breast tissue. Intelligent medical systems are expected to have substantial implications in healthcare politics by contributing to the diagnosis of indeterminate breast lesions by non-invasive imaging.

Learning, memory, and the role of neural network architecture.

PubMed

Hermundstad, Ann M; Brown, Kevin S; Bassett, Danielle S; Carlson, Jean M

2011-06-01

The performance of information processing systems, from artificial neural networks to natural neuronal ensembles, depends heavily on the underlying system architecture. In this study, we compare the performance of parallel and layered network architectures during sequential tasks that require both acquisition and retention of information, thereby identifying tradeoffs between learning and memory processes. During the task of supervised, sequential function approximation, networks produce and adapt representations of external information. Performance is evaluated by statistically analyzing the error in these representations while varying the initial network state, the structure of the external information, and the time given to learn the information. We link performance to complexity in network architecture by characterizing local error landscape curvature. We find that variations in error landscape structure give rise to tradeoffs in performance; these include the ability of the network to maximize accuracy versus minimize inaccuracy and produce specific versus generalizable representations of information. Parallel networks generate smooth error landscapes with deep, narrow minima, enabling them to find highly specific representations given sufficient time. While accurate, however, these representations are difficult to generalize. In contrast, layered networks generate rough error landscapes with a variety of local minima, allowing them to quickly find coarse representations. Although less accurate, these representations are easily adaptable. The presence of measurable performance tradeoffs in both layered and parallel networks has implications for understanding the behavior of a wide variety of natural and artificial learning systems.

Feature extraction for ultrasonic sensor based defect detection in ceramic components

NASA Astrophysics Data System (ADS)

Kesharaju, Manasa; Nagarajah, Romesh

2014-02-01

High density silicon carbide materials are commonly used as the ceramic element of hard armour inserts used in traditional body armour systems to reduce their weight, while providing improved hardness, strength and elastic response to stress. Currently, armour ceramic tiles are inspected visually offline using an X-ray technique that is time consuming and very expensive. In addition, from X-rays multiple defects are also misinterpreted as single defects. Therefore, to address these problems the ultrasonic non-destructive approach is being investigated. Ultrasound based inspection would be far more cost effective and reliable as the methodology is applicable for on-line quality control including implementation of accept/reject criteria. This paper describes a recently developed methodology to detect, locate and classify various manufacturing defects in ceramic tiles using sub band coding of ultrasonic test signals. The wavelet transform is applied to the ultrasonic signal and wavelet coefficients in the different frequency bands are extracted and used as input features to an artificial neural network (ANN) for purposes of signal classification. Two different classifiers, using artificial neural networks (supervised) and clustering (un-supervised) are supplied with features selected using Principal Component Analysis(PCA) and their classification performance compared. This investigation establishes experimentally that Principal Component Analysis(PCA) can be effectively used as a feature selection method that provides superior results for classifying various defects in the context of ultrasonic inspection in comparison with the X-ray technique.

Research on Improved Depth Belief Network-Based Prediction of Cardiovascular Diseases

PubMed Central

Zhang, Hongpo

2018-01-01

Quantitative analysis and prediction can help to reduce the risk of cardiovascular disease. Quantitative prediction based on traditional model has low accuracy. The variance of model prediction based on shallow neural network is larger. In this paper, cardiovascular disease prediction model based on improved deep belief network (DBN) is proposed. Using the reconstruction error, the network depth is determined independently, and unsupervised training and supervised optimization are combined. It ensures the accuracy of model prediction while guaranteeing stability. Thirty experiments were performed independently on the Statlog (Heart) and Heart Disease Database data sets in the UCI database. Experimental results showed that the mean of prediction accuracy was 91.26% and 89.78%, respectively. The variance of prediction accuracy was 5.78 and 4.46, respectively. PMID:29854369

Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning

PubMed Central

Hoo-Chang, Shin; Roth, Holger R.; Gao, Mingchen; Lu, Le; Xu, Ziyue; Nogues, Isabella; Yao, Jianhua; Mollura, Daniel

2016-01-01

Remarkable progress has been made in image recognition, primarily due to the availability of large-scale annotated datasets (i.e. ImageNet) and the revival of deep convolutional neural networks (CNN). CNNs enable learning data-driven, highly representative, layered hierarchical image features from sufficient training data. However, obtaining datasets as comprehensively annotated as ImageNet in the medical imaging domain remains a challenge. There are currently three major techniques that successfully employ CNNs to medical image classification: training the CNN from scratch, using off-the-shelf pre-trained CNN features, and conducting unsupervised CNN pre-training with supervised fine-tuning. Another effective method is transfer learning, i.e., fine-tuning CNN models (supervised) pre-trained from natural image dataset to medical image tasks (although domain transfer between two medical image datasets is also possible). In this paper, we exploit three important, but previously understudied factors of employing deep convolutional neural networks to computer-aided detection problems. We first explore and evaluate different CNN architectures. The studied models contain 5 thousand to 160 million parameters, and vary in numbers of layers. We then evaluate the influence of dataset scale and spatial image context on performance. Finally, we examine when and why transfer learning from pre-trained ImageNet (via fine-tuning) can be useful. We study two specific computeraided detection (CADe) problems, namely thoraco-abdominal lymph node (LN) detection and interstitial lung disease (ILD) classification. We achieve the state-of-the-art performance on the mediastinal LN detection, with 85% sensitivity at 3 false positive per patient, and report the first five-fold cross-validation classification results on predicting axial CT slices with ILD categories. Our extensive empirical evaluation, CNN model analysis and valuable insights can be extended to the design of high performance CAD systems for other medical imaging tasks. PMID:26886976

Video-based face recognition via convolutional neural networks

NASA Astrophysics Data System (ADS)

Bao, Tianlong; Ding, Chunhui; Karmoshi, Saleem; Zhu, Ming

2017-06-01

Face recognition has been widely studied recently while video-based face recognition still remains a challenging task because of the low quality and large intra-class variation of video captured face images. In this paper, we focus on two scenarios of video-based face recognition: 1)Still-to-Video(S2V) face recognition, i.e., querying a still face image against a gallery of video sequences; 2)Video-to-Still(V2S) face recognition, in contrast to S2V scenario. A novel method was proposed in this paper to transfer still and video face images to an Euclidean space by a carefully designed convolutional neural network, then Euclidean metrics are used to measure the distance between still and video images. Identities of still and video images that group as pairs are used as supervision. In the training stage, a joint loss function that measures the Euclidean distance between the predicted features of training pairs and expanding vectors of still images is optimized to minimize the intra-class variation while the inter-class variation is guaranteed due to the large margin of still images. Transferred features are finally learned via the designed convolutional neural network. Experiments are performed on COX face dataset. Experimental results show that our method achieves reliable performance compared with other state-of-the-art methods.

Radar detection with the Neyman-Pearson criterion using supervised-learning-machines trained with the cross-entropy error

NASA Astrophysics Data System (ADS)

Jarabo-Amores, María-Pilar; la Mata-Moya, David de; Gil-Pita, Roberto; Rosa-Zurera, Manuel

2013-12-01

The application of supervised learning machines trained to minimize the Cross-Entropy error to radar detection is explored in this article. The detector is implemented with a learning machine that implements a discriminant function, which output is compared to a threshold selected to fix a desired probability of false alarm. The study is based on the calculation of the function the learning machine approximates to during training, and the application of a sufficient condition for a discriminant function to be used to approximate the optimum Neyman-Pearson (NP) detector. In this article, the function a supervised learning machine approximates to after being trained to minimize the Cross-Entropy error is obtained. This discriminant function can be used to implement the NP detector, which maximizes the probability of detection, maintaining the probability of false alarm below or equal to a predefined value. Some experiments about signal detection using neural networks are also presented to test the validity of the study.

A neuromorphic network for generic multivariate data classification

PubMed Central

Schmuker, Michael; Pfeil, Thomas; Nawrot, Martin Paul

2014-01-01

Computational neuroscience has uncovered a number of computational principles used by nervous systems. At the same time, neuromorphic hardware has matured to a state where fast silicon implementations of complex neural networks have become feasible. En route to future technical applications of neuromorphic computing the current challenge lies in the identification and implementation of functional brain algorithms. Taking inspiration from the olfactory system of insects, we constructed a spiking neural network for the classification of multivariate data, a common problem in signal and data analysis. In this model, real-valued multivariate data are converted into spike trains using “virtual receptors” (VRs). Their output is processed by lateral inhibition and drives a winner-take-all circuit that supports supervised learning. VRs are conveniently implemented in software, whereas the lateral inhibition and classification stages run on accelerated neuromorphic hardware. When trained and tested on real-world datasets, we find that the classification performance is on par with a naïve Bayes classifier. An analysis of the network dynamics shows that stable decisions in output neuron populations are reached within less than 100 ms of biological time, matching the time-to-decision reported for the insect nervous system. Through leveraging a population code, the network tolerates the variability of neuronal transfer functions and trial-to-trial variation that is inevitably present on the hardware system. Our work provides a proof of principle for the successful implementation of a functional spiking neural network on a configurable neuromorphic hardware system that can readily be applied to real-world computing problems. PMID:24469794

Spectral Classification of Galaxies at 0.5 <= z <= 1 in the CDFS: The Artificial Neural Network Approach

NASA Astrophysics Data System (ADS)

Teimoorinia, H.

2012-12-01

The aim of this work is to combine spectral energy distribution (SED) fitting with artificial neural network techniques to assign spectral characteristics to a sample of galaxies at 0.5 < z < 1. The sample is selected from the spectroscopic campaign of the ESO/GOODS-South field, with 142 sources having photometric data from the GOODS-MUSIC catalog covering bands between ~0.4 and 24 μm in 10-13 filters. We use the CIGALE code to fit photometric data to Maraston's synthesis spectra to derive mass, specific star formation rate, and age, as well as the best SED of the galaxies. We use the spectral models presented by Kinney et al. as targets in the wavelength interval ~1200-7500 Å. Then a series of neural networks are trained, with average performance ~90%, to classify the best SED in a supervised manner. We consider the effects of the prominent features of the best SED on the performance of the trained networks and also test networks on the galaxy spectra of Coleman et al., which have a lower resolution than the target models. In this way, we conclude that the trained networks take into account all the features of the spectra simultaneously. Using the method, 105 out of 142 galaxies of the sample are classified with high significance. The locus of the classified galaxies in the three graphs of the physical parameters of mass, age, and specific star formation rate appears consistent with the morphological characteristics of the galaxies.

SPECTRAL CLASSIFICATION OF GALAXIES AT 0.5 {<=} z {<=} 1 IN THE CDFS: THE ARTIFICIAL NEURAL NETWORK APPROACH

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

Teimoorinia, H., E-mail: hteimoo@uvic.ca

2012-12-01

The aim of this work is to combine spectral energy distribution (SED) fitting with artificial neural network techniques to assign spectral characteristics to a sample of galaxies at 0.5 < z < 1. The sample is selected from the spectroscopic campaign of the ESO/GOODS-South field, with 142 sources having photometric data from the GOODS-MUSIC catalog covering bands between {approx}0.4 and 24 {mu}m in 10-13 filters. We use the CIGALE code to fit photometric data to Maraston's synthesis spectra to derive mass, specific star formation rate, and age, as well as the best SED of the galaxies. We use the spectralmore » models presented by Kinney et al. as targets in the wavelength interval {approx}1200-7500 A. Then a series of neural networks are trained, with average performance {approx}90%, to classify the best SED in a supervised manner. We consider the effects of the prominent features of the best SED on the performance of the trained networks and also test networks on the galaxy spectra of Coleman et al., which have a lower resolution than the target models. In this way, we conclude that the trained networks take into account all the features of the spectra simultaneously. Using the method, 105 out of 142 galaxies of the sample are classified with high significance. The locus of the classified galaxies in the three graphs of the physical parameters of mass, age, and specific star formation rate appears consistent with the morphological characteristics of the galaxies.« less

Virus Particle Detection by Convolutional Neural Network in Transmission Electron Microscopy Images.

PubMed

Ito, Eisuke; Sato, Takaaki; Sano, Daisuke; Utagawa, Etsuko; Kato, Tsuyoshi

2018-06-01

A new computational method for the detection of virus particles in transmission electron microscopy (TEM) images is presented. Our approach is to use a convolutional neural network that transforms a TEM image to a probabilistic map that indicates where virus particles exist in the image. Our proposed approach automatically and simultaneously learns both discriminative features and classifier for virus particle detection by machine learning, in contrast to existing methods that are based on handcrafted features that yield many false positives and require several postprocessing steps. The detection performance of the proposed method was assessed against a dataset of TEM images containing feline calicivirus particles and compared with several existing detection methods, and the state-of-the-art performance of the developed method for detecting virus was demonstrated. Since our method is based on supervised learning that requires both the input images and their corresponding annotations, it is basically used for detection of already-known viruses. However, the method is highly flexible, and the convolutional networks can adapt themselves to any virus particles by learning automatically from an annotated dataset.

Deep Learning for Drug Design: an Artificial Intelligence Paradigm for Drug Discovery in the Big Data Era.

PubMed

Jing, Yankang; Bian, Yuemin; Hu, Ziheng; Wang, Lirong; Xie, Xiang-Qun Sean

2018-03-30

Over the last decade, deep learning (DL) methods have been extremely successful and widely used to develop artificial intelligence (AI) in almost every domain, especially after it achieved its proud record on computational Go. Compared to traditional machine learning (ML) algorithms, DL methods still have a long way to go to achieve recognition in small molecular drug discovery and development. And there is still lots of work to do for the popularization and application of DL for research purpose, e.g., for small molecule drug research and development. In this review, we mainly discussed several most powerful and mainstream architectures, including the convolutional neural network (CNN), recurrent neural network (RNN), and deep auto-encoder networks (DAENs), for supervised learning and nonsupervised learning; summarized most of the representative applications in small molecule drug design; and briefly introduced how DL methods were used in those applications. The discussion for the pros and cons of DL methods as well as the main challenges we need to tackle were also emphasized.

Bidirectional RNN for Medical Event Detection in Electronic Health Records.

PubMed

Jagannatha, Abhyuday N; Yu, Hong

2016-06-01

Sequence labeling for extraction of medical events and their attributes from unstructured text in Electronic Health Record (EHR) notes is a key step towards semantic understanding of EHRs. It has important applications in health informatics including pharmacovigilance and drug surveillance. The state of the art supervised machine learning models in this domain are based on Conditional Random Fields (CRFs) with features calculated from fixed context windows. In this application, we explored recurrent neural network frameworks and show that they significantly out-performed the CRF models.

Multi-Level and Multi-Scale Feature Aggregation Using Pretrained Convolutional Neural Networks for Music Auto-Tagging

NASA Astrophysics Data System (ADS)

Lee, Jongpil; Nam, Juhan

2017-08-01

Music auto-tagging is often handled in a similar manner to image classification by regarding the 2D audio spectrogram as image data. However, music auto-tagging is distinguished from image classification in that the tags are highly diverse and have different levels of abstractions. Considering this issue, we propose a convolutional neural networks (CNN)-based architecture that embraces multi-level and multi-scaled features. The architecture is trained in three steps. First, we conduct supervised feature learning to capture local audio features using a set of CNNs with different input sizes. Second, we extract audio features from each layer of the pre-trained convolutional networks separately and aggregate them altogether given a long audio clip. Finally, we put them into fully-connected networks and make final predictions of the tags. Our experiments show that using the combination of multi-level and multi-scale features is highly effective in music auto-tagging and the proposed method outperforms previous state-of-the-arts on the MagnaTagATune dataset and the Million Song Dataset. We further show that the proposed architecture is useful in transfer learning.

A recurrent neural network for classification of unevenly sampled variable stars

NASA Astrophysics Data System (ADS)

Naul, Brett; Bloom, Joshua S.; Pérez, Fernando; van der Walt, Stéfan

2018-02-01

Astronomical surveys of celestial sources produce streams of noisy time series measuring flux versus time (`light curves'). Unlike in many other physical domains, however, large (and source-specific) temporal gaps in data arise naturally due to intranight cadence choices as well as diurnal and seasonal constraints1-5. With nightly observations of millions of variable stars and transients from upcoming surveys4,6, efficient and accurate discovery and classification techniques on noisy, irregularly sampled data must be employed with minimal human-in-the-loop involvement. Machine learning for inference tasks on such data traditionally requires the laborious hand-coding of domain-specific numerical summaries of raw data (`features')7. Here, we present a novel unsupervised autoencoding recurrent neural network8 that makes explicit use of sampling times and known heteroskedastic noise properties. When trained on optical variable star catalogues, this network produces supervised classification models that rival other best-in-class approaches. We find that autoencoded features learned in one time-domain survey perform nearly as well when applied to another survey. These networks can continue to learn from new unlabelled observations and may be used in other unsupervised tasks, such as forecasting and anomaly detection.

A comparative evaluation of supervised and unsupervised representation learning approaches for anaplastic medulloblastoma differentiation

NASA Astrophysics Data System (ADS)

Cruz-Roa, Angel; Arevalo, John; Basavanhally, Ajay; Madabhushi, Anant; González, Fabio

2015-01-01

Learning data representations directly from the data itself is an approach that has shown great success in different pattern recognition problems, outperforming state-of-the-art feature extraction schemes for different tasks in computer vision, speech recognition and natural language processing. Representation learning applies unsupervised and supervised machine learning methods to large amounts of data to find building-blocks that better represent the information in it. Digitized histopathology images represents a very good testbed for representation learning since it involves large amounts of high complex, visual data. This paper presents a comparative evaluation of different supervised and unsupervised representation learning architectures to specifically address open questions on what type of learning architectures (deep or shallow), type of learning (unsupervised or supervised) is optimal. In this paper we limit ourselves to addressing these questions in the context of distinguishing between anaplastic and non-anaplastic medulloblastomas from routine haematoxylin and eosin stained images. The unsupervised approaches evaluated were sparse autoencoders and topographic reconstruct independent component analysis, and the supervised approach was convolutional neural networks. Experimental results show that shallow architectures with more neurons are better than deeper architectures without taking into account local space invariances and that topographic constraints provide useful invariant features in scale and rotations for efficient tumor differentiation.

Looking for underlying features in automatic and reviewed seismic bulletins through a neural network

NASA Astrophysics Data System (ADS)

Carluccio, R.; Console, R.; Chiappini, M.; Chiappini, S.

2009-12-01

SEL1 bulletins are, among all IDC products, a fundamental tool for NDCs in their task of national assessment of compliance with the CTBT. This is because SEL1s are expected to be disseminated within 2 hours from the occurrence of any detected waveform event, and the National Authorities are supposed to take a political decision in nearly real time, especially in the case when the event could triggers the request for an on site inspection. In this context not only the rapidity, but also the reliability of the SEL1 is a fundamental requirement. Our last years experience gained in the comparison between SEL1 and Italian Seismic Bulletin events has shown that SEL1s usually contain a big fraction of bogus events (sometimes close to 50%). This is due to many factors, all related to the availability of processing data and to the fast automatic algorithms involved. On the other hand, REBs are much more reliable as proved by our experience. Therefore, in spite of their relevant time delay by which they are distributed, which prevents their real-time use, REBs can be still useful in a retrospective way as reference information for comparison with SEL1s. This study tries to set up a sort of logical filter on the SEL1s that, while maintaining the rapidity requirements, improves their reliability. Our idea is based on the assumption that the SEL1s are produced by systematic algorithm of phase association and therefore some patterns among the input and output data could exist and be recognized. Our approach was initially based on a set of rules suggested by human experts on their personal experience, and its application on large datasets on a global scale. Other approaches not involving human interaction (data mining techniques) do exist. This study refers specifically to a semi-automatic approach: fitting of multi-parametric relationships hidden in the data set, through the application of neural networks by an algorithm of supervised learning. Full SEL1 and REB bulletins from Jan 2005 to Oct 2008 have been inserted in a database, together with IMS stations availability information. Part of these data have been used to create two sets of independent data (learning and verifying) used to train a "feed-forward" supervised neural network. A network supervised training algorithm using "confirmation flag" values has been used. In order to optimize network training input a significant, not redundant subset of input parameters has been looked for with the help of a genetic algorithm search tool. A suitable 12 input subset has been found and a network architecture of 12-20-1 has thus been chosen and trained on a 15094 records data set. Different runs of training sequences have been conducted, all showing CCR (Correct Classification Rate) values of the order of 75% - 80%. The trained network behavior is shown in term of ROC curve and input-out success-error matrices. The results of the analysis on our testing and validating data groups appear promising.

Physical Realization of a Supervised Learning System Built with Organic Memristive Synapses

NASA Astrophysics Data System (ADS)

Lin, Yu-Pu; Bennett, Christopher H.; Cabaret, Théo; Vodenicarevic, Damir; Chabi, Djaafar; Querlioz, Damien; Jousselme, Bruno; Derycke, Vincent; Klein, Jacques-Olivier

2016-09-01

Multiple modern applications of electronics call for inexpensive chips that can perform complex operations on natural data with limited energy. A vision for accomplishing this is implementing hardware neural networks, which fuse computation and memory, with low cost organic electronics. A challenge, however, is the implementation of synapses (analog memories) composed of such materials. In this work, we introduce robust, fastly programmable, nonvolatile organic memristive nanodevices based on electrografted redox complexes that implement synapses thanks to a wide range of accessible intermediate conductivity states. We demonstrate experimentally an elementary neural network, capable of learning functions, which combines four pairs of organic memristors as synapses and conventional electronics as neurons. Our architecture is highly resilient to issues caused by imperfect devices. It tolerates inter-device variability and an adaptable learning rule offers immunity against asymmetries in device switching. Highly compliant with conventional fabrication processes, the system can be extended to larger computing systems capable of complex cognitive tasks, as demonstrated in complementary simulations.

Automotive System for Remote Surface Classification.

PubMed

Bystrov, Aleksandr; Hoare, Edward; Tran, Thuy-Yung; Clarke, Nigel; Gashinova, Marina; Cherniakov, Mikhail

2017-04-01

In this paper we shall discuss a novel approach to road surface recognition, based on the analysis of backscattered microwave and ultrasonic signals. The novelty of our method is sonar and polarimetric radar data fusion, extraction of features for separate swathes of illuminated surface (segmentation), and using of multi-stage artificial neural network for surface classification. The developed system consists of 24 GHz radar and 40 kHz ultrasonic sensor. The features are extracted from backscattered signals and then the procedures of principal component analysis and supervised classification are applied to feature data. The special attention is paid to multi-stage artificial neural network which allows an overall increase in classification accuracy. The proposed technique was tested for recognition of a large number of real surfaces in different weather conditions with the average accuracy of correct classification of 95%. The obtained results thereby demonstrate that the use of proposed system architecture and statistical methods allow for reliable discrimination of various road surfaces in real conditions.

Automotive System for Remote Surface Classification

PubMed Central

Bystrov, Aleksandr; Hoare, Edward; Tran, Thuy-Yung; Clarke, Nigel; Gashinova, Marina; Cherniakov, Mikhail

2017-01-01

In this paper we shall discuss a novel approach to road surface recognition, based on the analysis of backscattered microwave and ultrasonic signals. The novelty of our method is sonar and polarimetric radar data fusion, extraction of features for separate swathes of illuminated surface (segmentation), and using of multi-stage artificial neural network for surface classification. The developed system consists of 24 GHz radar and 40 kHz ultrasonic sensor. The features are extracted from backscattered signals and then the procedures of principal component analysis and supervised classification are applied to feature data. The special attention is paid to multi-stage artificial neural network which allows an overall increase in classification accuracy. The proposed technique was tested for recognition of a large number of real surfaces in different weather conditions with the average accuracy of correct classification of 95%. The obtained results thereby demonstrate that the use of proposed system architecture and statistical methods allow for reliable discrimination of various road surfaces in real conditions. PMID:28368297

Neural Network Training by Integration of Adjoint Systems of Equations Forward in Time

NASA Technical Reports Server (NTRS)

Toomarian, Nikzad (Inventor); Barhen, Jacob (Inventor)

1999-01-01

A method and apparatus for supervised neural learning of time dependent trajectories exploits the concepts of adjoint operators to enable computation of the gradient of an objective functional with respect to the various parameters of the network architecture in a highly efficient manner. Specifically. it combines the advantage of dramatic reductions in computational complexity inherent in adjoint methods with the ability to solve two adjoint systems of equations together forward in time. Not only is a large amount of computation and storage saved. but the handling of real-time applications becomes also possible. The invention has been applied it to two examples of representative complexity which have recently been analyzed in the open literature and demonstrated that a circular trajectory can be learned in approximately 200 iterations compared to the 12000 reported in the literature. A figure eight trajectory was achieved in under 500 iterations compared to 20000 previously required. Tbc trajectories computed using our new method are much closer to the target trajectories than was reported in previous studies.

Neural network training by integration of adjoint systems of equations forward in time

NASA Technical Reports Server (NTRS)

Toomarian, Nikzad (Inventor); Barhen, Jacob (Inventor)

1992-01-01

A method and apparatus for supervised neural learning of time dependent trajectories exploits the concepts of adjoint operators to enable computation of the gradient of an objective functional with respect to the various parameters of the network architecture in a highly efficient manner. Specifically, it combines the advantage of dramatic reductions in computational complexity inherent in adjoint methods with the ability to solve two adjoint systems of equations together forward in time. Not only is a large amount of computation and storage saved, but the handling of real-time applications becomes also possible. The invention has been applied it to two examples of representative complexity which have recently been analyzed in the open literature and demonstrated that a circular trajectory can be learned in approximately 200 iterations compared to the 12000 reported in the literature. A figure eight trajectory was achieved in under 500 iterations compared to 20000 previously required. The trajectories computed using our new method are much closer to the target trajectories than was reported in previous studies.

A new simple /spl infin/OH neuron model as a biologically plausible principal component analyzer.

PubMed

Jankovic, M V

2003-01-01

A new approach to unsupervised learning in a single-layer neural network is discussed. An algorithm for unsupervised learning based upon the Hebbian learning rule is presented. A simple neuron model is analyzed. A dynamic neural model, which contains both feed-forward and feedback connections between the input and the output, has been adopted. The, proposed learning algorithm could be more correctly named self-supervised rather than unsupervised. The solution proposed here is a modified Hebbian rule, in which the modification of the synaptic strength is proportional not to pre- and postsynaptic activity, but instead to the presynaptic and averaged value of postsynaptic activity. It is shown that the model neuron tends to extract the principal component from a stationary input vector sequence. Usually accepted additional decaying terms for the stabilization of the original Hebbian rule are avoided. Implementation of the basic Hebbian scheme would not lead to unrealistic growth of the synaptic strengths, thanks to the adopted network structure.

Learning disordered topological phases by statistical recovery of symmetry

NASA Astrophysics Data System (ADS)

Yoshioka, Nobuyuki; Akagi, Yutaka; Katsura, Hosho

2018-05-01

We apply the artificial neural network in a supervised manner to map out the quantum phase diagram of disordered topological superconductors in class DIII. Given the disorder that keeps the discrete symmetries of the ensemble as a whole, translational symmetry which is broken in the quasiparticle distribution individually is recovered statistically by taking an ensemble average. By using this, we classify the phases by the artificial neural network that learned the quasiparticle distribution in the clean limit and show that the result is totally consistent with the calculation by the transfer matrix method or noncommutative geometry approach. If all three phases, namely the Z2, trivial, and thermal metal phases, appear in the clean limit, the machine can classify them with high confidence over the entire phase diagram. If only the former two phases are present, we find that the machine remains confused in a certain region, leading us to conclude the detection of the unknown phase which is eventually identified as the thermal metal phase.

Physical Realization of a Supervised Learning System Built with Organic Memristive Synapses.

PubMed

Lin, Yu-Pu; Bennett, Christopher H; Cabaret, Théo; Vodenicarevic, Damir; Chabi, Djaafar; Querlioz, Damien; Jousselme, Bruno; Derycke, Vincent; Klein, Jacques-Olivier

2016-09-07

Multiple modern applications of electronics call for inexpensive chips that can perform complex operations on natural data with limited energy. A vision for accomplishing this is implementing hardware neural networks, which fuse computation and memory, with low cost organic electronics. A challenge, however, is the implementation of synapses (analog memories) composed of such materials. In this work, we introduce robust, fastly programmable, nonvolatile organic memristive nanodevices based on electrografted redox complexes that implement synapses thanks to a wide range of accessible intermediate conductivity states. We demonstrate experimentally an elementary neural network, capable of learning functions, which combines four pairs of organic memristors as synapses and conventional electronics as neurons. Our architecture is highly resilient to issues caused by imperfect devices. It tolerates inter-device variability and an adaptable learning rule offers immunity against asymmetries in device switching. Highly compliant with conventional fabrication processes, the system can be extended to larger computing systems capable of complex cognitive tasks, as demonstrated in complementary simulations.

Applying deep neural networks to HEP job classification

NASA Astrophysics Data System (ADS)

Wang, L.; Shi, J.; Yan, X.

2015-12-01

The cluster of IHEP computing center is a middle-sized computing system which provides 10 thousands CPU cores, 5 PB disk storage, and 40 GB/s IO throughput. Its 1000+ users come from a variety of HEP experiments. In such a system, job classification is an indispensable task. Although experienced administrator can classify a HEP job by its IO pattern, it is unpractical to classify millions of jobs manually. We present how to solve this problem with deep neural networks in a supervised learning way. Firstly, we built a training data set of 320K samples by an IO pattern collection agent and a semi-automatic process of sample labelling. Then we implemented and trained DNNs models with Torch. During the process of model training, several meta-parameters was tuned with cross-validations. Test results show that a 5- hidden-layer DNNs model achieves 96% precision on the classification task. By comparison, it outperforms a linear model by 8% precision.

Surface inspection of flat products by means of texture analysis: on-line implementation using neural networks

NASA Astrophysics Data System (ADS)

Fernandez, Carlos; Platero, Carlos; Campoy, Pascual; Aracil, Rafael

1994-11-01

This paper describes some texture-based techniques that can be applied to quality assessment of flat products continuously produced (metal strips, wooden surfaces, cork, textile products, ...). Since the most difficult task is that of inspecting for product appearance, human-like inspection ability is required. A common feature to all these products is the presence of non- deterministic texture on their surfaces. Two main subjects are discussed: statistical techniques for both surface finishing determination and surface defect analysis as well as real-time implementation for on-line inspection in high-speed applications. For surface finishing determination a Gray Level Difference technique is presented to perform over low resolution images, that is, no-zoomed images. Defect analysis is performed by means of statistical texture analysis over defective portions of the surface. On-line implementation is accomplished by means of neural networks. When a defect arises, textural analysis is applied which result in a data-vector, acting as input of a neural net, previously trained in a supervised way. This approach tries to reach on-line performance in automated visual inspection applications when texture is presented in flat product surfaces.

Fine-Tuning Neural Patient Question Retrieval Model with Generative Adversarial Networks.

PubMed

Tang, Guoyu; Ni, Yuan; Wang, Keqiang; Yong, Qin

2018-01-01

The online patient question and answering (Q&A) system attracts an increasing amount of users in China. Patient will post their questions and wait for doctors' response. To avoid the lag time involved with the waiting and to reduce the workload on the doctors, a better method is to automatically retrieve the semantically equivalent question from the archive. We present a Generative Adversarial Networks (GAN) based approach to automatically retrieve patient question. We apply supervised deep learning based approaches to determine the similarity between patient questions. Then a GAN framework is used to fine-tune the pre-trained deep learning models. The experiment results show that fine-tuning by GAN can improve the performance.

Vision-based mobile robot navigation through deep convolutional neural networks and end-to-end learning

NASA Astrophysics Data System (ADS)

Zhang, Yachu; Zhao, Yuejin; Liu, Ming; Dong, Liquan; Kong, Lingqin; Liu, Lingling

2017-09-01

In contrast to humans, who use only visual information for navigation, many mobile robots use laser scanners and ultrasonic sensors along with vision cameras to navigate. This work proposes a vision-based robot control algorithm based on deep convolutional neural networks. We create a large 15-layer convolutional neural network learning system and achieve the advanced recognition performance. Our system is trained from end to end to map raw input images to direction in supervised mode. The images of data sets are collected in a wide variety of weather conditions and lighting conditions. Besides, the data sets are augmented by adding Gaussian noise and Salt-and-pepper noise to avoid overfitting. The algorithm is verified by two experiments, which are line tracking and obstacle avoidance. The line tracking experiment is proceeded in order to track the desired path which is composed of straight and curved lines. The goal of obstacle avoidance experiment is to avoid the obstacles indoor. Finally, we get 3.29% error rate on the training set and 5.1% error rate on the test set in the line tracking experiment, 1.8% error rate on the training set and less than 5% error rate on the test set in the obstacle avoidance experiment. During the actual test, the robot can follow the runway centerline outdoor and avoid the obstacle in the room accurately. The result confirms the effectiveness of the algorithm and our improvement in the network structure and train parameters

An Efficient Supervised Training Algorithm for Multilayer Spiking Neural Networks

PubMed Central

Xie, Xiurui; Qu, Hong; Liu, Guisong; Zhang, Malu; Kurths, Jürgen

2016-01-01

The spiking neural networks (SNNs) are the third generation of neural networks and perform remarkably well in cognitive tasks such as pattern recognition. The spike emitting and information processing mechanisms found in biological cognitive systems motivate the application of the hierarchical structure and temporal encoding mechanism in spiking neural networks, which have exhibited strong computational capability. However, the hierarchical structure and temporal encoding approach require neurons to process information serially in space and time respectively, which reduce the training efficiency significantly. For training the hierarchical SNNs, most existing methods are based on the traditional back-propagation algorithm, inheriting its drawbacks of the gradient diffusion and the sensitivity on parameters. To keep the powerful computation capability of the hierarchical structure and temporal encoding mechanism, but to overcome the low efficiency of the existing algorithms, a new training algorithm, the Normalized Spiking Error Back Propagation (NSEBP) is proposed in this paper. In the feedforward calculation, the output spike times are calculated by solving the quadratic function in the spike response model instead of detecting postsynaptic voltage states at all time points in traditional algorithms. Besides, in the feedback weight modification, the computational error is propagated to previous layers by the presynaptic spike jitter instead of the gradient decent rule, which realizes the layer-wised training. Furthermore, our algorithm investigates the mathematical relation between the weight variation and voltage error change, which makes the normalization in the weight modification applicable. Adopting these strategies, our algorithm outperforms the traditional SNN multi-layer algorithms in terms of learning efficiency and parameter sensitivity, that are also demonstrated by the comprehensive experimental results in this paper. PMID:27044001

Implementations of back propagation algorithm in ecosystems applications

NASA Astrophysics Data System (ADS)

Ali, Khalda F.; Sulaiman, Riza; Elamir, Amir Mohamed

2015-05-01

Artificial Neural Networks (ANNs) have been applied to an increasing number of real world problems of considerable complexity. Their most important advantage is in solving problems which are too complex for conventional technologies, that do not have an algorithmic solutions or their algorithmic Solutions is too complex to be found. In general, because of their abstraction from the biological brain, ANNs are developed from concept that evolved in the late twentieth century neuro-physiological experiments on the cells of the human brain to overcome the perceived inadequacies with conventional ecological data analysis methods. ANNs have gained increasing attention in ecosystems applications, because of ANN's capacity to detect patterns in data through non-linear relationships, this characteristic confers them a superior predictive ability. In this research, ANNs is applied in an ecological system analysis. The neural networks use the well known Back Propagation (BP) Algorithm with the Delta Rule for adaptation of the system. The Back Propagation (BP) training Algorithm is an effective analytical method for adaptation of the ecosystems applications, the main reason because of their capacity to detect patterns in data through non-linear relationships. This characteristic confers them a superior predicting ability. The BP algorithm uses supervised learning, which means that we provide the algorithm with examples of the inputs and outputs we want the network to compute, and then the error is calculated. The idea of the back propagation algorithm is to reduce this error, until the ANNs learns the training data. The training begins with random weights, and the goal is to adjust them so that the error will be minimal. This research evaluated the use of artificial neural networks (ANNs) techniques in an ecological system analysis and modeling. The experimental results from this research demonstrate that an artificial neural network system can be trained to act as an expert ecosystem analyzer for many applications in ecological fields. The pilot ecosystem analyzer shows promising ability for generalization and requires further tuning and refinement of the basis neural network system for optimal performance.

Mineral and Geochemical Classification From Spectroscopy/Diffraction Through Neural Networks

NASA Astrophysics Data System (ADS)

Ferralis, N.; Grossman, J.; Summons, R. E.

2017-12-01

Spectroscopy and diffraction techniques are essential for understanding structural, chemical and functional properties of geological materials for Earth and Planetary Sciences. Beyond data collection, quantitative insight relies on experimentally assembled, or computationally derived spectra. Inference on the geochemical or geophysical properties (such as crystallographic order, chemical functionality, elemental composition, etc.) of a particular geological material (mineral, organic matter, etc.) is based on fitting unknown spectra and comparing the fit with consolidated databases. The complexity of fitting highly convoluted spectra, often limits the ability to infer geochemical characteristics, and limits the throughput for extensive datasets. With the emergence of heuristic approaches to pattern recognitions though machine learning, in this work we investigate the possibility and potential of using supervised neural networks trained on available public spectroscopic database to directly infer geochemical parameters from unknown spectra. Using Raman, infrared spectroscopy and powder x-ray diffraction from the publicly available RRUFF database, we train neural network models to classify mineral and organic compounds (pure or mixtures) based on crystallographic structure from diffraction, chemical functionality, elemental composition and bonding from spectroscopy. As expected, the accuracy of the inference is strongly dependent on the quality and extent of the training data. We will identify a series of requirements and guidelines for the training dataset needed to achieve consistent high accuracy inference, along with methods to compensate for limited of data.

Authorship attribution of source code by using back propagation neural network based on particle swarm optimization

PubMed Central

Xu, Guoai; Li, Qi; Guo, Yanhui; Zhang, Miao

2017-01-01

Authorship attribution is to identify the most likely author of a given sample among a set of candidate known authors. It can be not only applied to discover the original author of plain text, such as novels, blogs, emails, posts etc., but also used to identify source code programmers. Authorship attribution of source code is required in diverse applications, ranging from malicious code tracking to solving authorship dispute or software plagiarism detection. This paper aims to propose a new method to identify the programmer of Java source code samples with a higher accuracy. To this end, it first introduces back propagation (BP) neural network based on particle swarm optimization (PSO) into authorship attribution of source code. It begins by computing a set of defined feature metrics, including lexical and layout metrics, structure and syntax metrics, totally 19 dimensions. Then these metrics are input to neural network for supervised learning, the weights of which are output by PSO and BP hybrid algorithm. The effectiveness of the proposed method is evaluated on a collected dataset with 3,022 Java files belong to 40 authors. Experiment results show that the proposed method achieves 91.060% accuracy. And a comparison with previous work on authorship attribution of source code for Java language illustrates that this proposed method outperforms others overall, also with an acceptable overhead. PMID:29095934

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

AllamehZadeh, Mostafa, E-mail: dibaparima@yahoo.com

A Quadratic Neural Networks (QNNs) model has been developed for identifying seismic source classification problem at regional distances using ARMA coefficients determination by Artificial Neural Networks (ANNs). We have devised a supervised neural system to discriminate between earthquakes and chemical explosions with filter coefficients obtained by windowed P-wave phase spectra (15 s). First, we preprocess the recording's signals to cancel out instrumental and attenuation site effects and obtain a compact representation of seismic records. Second, we use a QNNs system to obtain ARMA coefficients for feature extraction in the discrimination problem. The derived coefficients are then applied to the neuralmore » system to train and classification. In this study, we explore the possibility of using single station three-component (3C) covariance matrix traces from a priori-known explosion sites (learning) for automatically recognizing subsequent explosions from the same site. The results have shown that this feature extraction gives the best classifier for seismic signals and performs significantly better than other classification methods. The events have been tested, which include 36 chemical explosions at the Semipalatinsk test site in Kazakhstan and 61 earthquakes (mb = 5.0-6.5) recorded by the Iranian National Seismic Network (INSN). The 100% correct decisions were obtained between site explosions and some of non-site events. The above approach to event discrimination is very flexible as we can combine several 3C stations.« less

Deep nets vs expert designed features in medical physics: An IMRT QA case study.

PubMed

Interian, Yannet; Rideout, Vincent; Kearney, Vasant P; Gennatas, Efstathios; Morin, Olivier; Cheung, Joey; Solberg, Timothy; Valdes, Gilmer

2018-03-30

The purpose of this study was to compare the performance of Deep Neural Networks against a technique designed by domain experts in the prediction of gamma passing rates for Intensity Modulated Radiation Therapy Quality Assurance (IMRT QA). A total of 498 IMRT plans across all treatment sites were planned in Eclipse version 11 and delivered using a dynamic sliding window technique on Clinac iX or TrueBeam Linacs. Measurements were performed using a commercial 2D diode array, and passing rates for 3%/3 mm local dose/distance-to-agreement (DTA) were recorded. Separately, fluence maps calculated for each plan were used as inputs to a convolution neural network (CNN). The CNNs were trained to predict IMRT QA gamma passing rates using TensorFlow and Keras. A set of model architectures, inspired by the convolutional blocks of the VGG-16 ImageNet model, were constructed and implemented. Synthetic data, created by rotating and translating the fluence maps during training, was created to boost the performance of the CNNs. Dropout, batch normalization, and data augmentation were utilized to help train the model. The performance of the CNNs was compared to a generalized Poisson regression model, previously developed for this application, which used 78 expert designed features. Deep Neural Networks without domain knowledge achieved comparable performance to a baseline system designed by domain experts in the prediction of 3%/3 mm Local gamma passing rates. An ensemble of neural nets resulted in a mean absolute error (MAE) of 0.70 ± 0.05 and the domain expert model resulted in a 0.74 ± 0.06. Convolutional neural networks (CNNs) with transfer learning can predict IMRT QA passing rates by automatically designing features from the fluence maps without human expert supervision. Predictions from CNNs are comparable to a system carefully designed by physicist experts. © 2018 American Association of Physicists in Medicine.

Identifying images of handwritten digits using deep learning in H2O

NASA Astrophysics Data System (ADS)

Sadhasivam, Jayakumar; Charanya, R.; Kumar, S. Harish; Srinivasan, A.

2017-11-01

Automatic digit recognition is of popular interest today. Deep learning techniques make it possible for object recognition in image data. Perceiving the digit has turned into a fundamental part as far as certifiable applications. Since, digits are composed in various styles in this way to distinguish the digit it is important to perceive and arrange it with the assistance of machine learning methods. This exploration depends on supervised learning vector quantization neural system arranged under counterfeit artificial neural network. The pictures of digits are perceived, prepared and tried. After the system is made digits are prepared utilizing preparing dataset vectors and testing is connected to the pictures of digits which are separated to each other by fragmenting the picture and resizing the digit picture as needs be for better precision.

Identifying QCD Transition Using Deep Learning

NASA Astrophysics Data System (ADS)

Zhou, Kai; Pang, Long-gang; Su, Nan; Petersen, Hannah; Stoecker, Horst; Wang, Xin-Nian

2018-02-01

In this proceeding we review our recent work using supervised learning with a deep convolutional neural network (CNN) to identify the QCD equation of state (EoS) employed in hydrodynamic modeling of heavy-ion collisions given only final-state particle spectra ρ(pT, V). We showed that there is a traceable encoder of the dynamical information from phase structure (EoS) that survives the evolution and exists in the final snapshot, which enables the trained CNN to act as an effective "EoS-meter" in detecting the nature of the QCD transition.

Evolutionary Design of Convolutional Neural Networks for Human Activity Recognition in Sensor-Rich Environments.

PubMed

Baldominos, Alejandro; Saez, Yago; Isasi, Pedro

2018-04-23

Human activity recognition is a challenging problem for context-aware systems and applications. It is gaining interest due to the ubiquity of different sensor sources, wearable smart objects, ambient sensors, etc. This task is usually approached as a supervised machine learning problem, where a label is to be predicted given some input data, such as the signals retrieved from different sensors. For tackling the human activity recognition problem in sensor network environments, in this paper we propose the use of deep learning (convolutional neural networks) to perform activity recognition using the publicly available OPPORTUNITY dataset. Instead of manually choosing a suitable topology, we will let an evolutionary algorithm design the optimal topology in order to maximize the classification F1 score. After that, we will also explore the performance of committees of the models resulting from the evolutionary process. Results analysis indicates that the proposed model was able to perform activity recognition within a heterogeneous sensor network environment, achieving very high accuracies when tested with new sensor data. Based on all conducted experiments, the proposed neuroevolutionary system has proved to be able to systematically find a classification model which is capable of outperforming previous results reported in the state-of-the-art, showing that this approach is useful and improves upon previously manually-designed architectures.

Evolutionary Design of Convolutional Neural Networks for Human Activity Recognition in Sensor-Rich Environments

PubMed Central

2018-01-01

Human activity recognition is a challenging problem for context-aware systems and applications. It is gaining interest due to the ubiquity of different sensor sources, wearable smart objects, ambient sensors, etc. This task is usually approached as a supervised machine learning problem, where a label is to be predicted given some input data, such as the signals retrieved from different sensors. For tackling the human activity recognition problem in sensor network environments, in this paper we propose the use of deep learning (convolutional neural networks) to perform activity recognition using the publicly available OPPORTUNITY dataset. Instead of manually choosing a suitable topology, we will let an evolutionary algorithm design the optimal topology in order to maximize the classification F1 score. After that, we will also explore the performance of committees of the models resulting from the evolutionary process. Results analysis indicates that the proposed model was able to perform activity recognition within a heterogeneous sensor network environment, achieving very high accuracies when tested with new sensor data. Based on all conducted experiments, the proposed neuroevolutionary system has proved to be able to systematically find a classification model which is capable of outperforming previous results reported in the state-of-the-art, showing that this approach is useful and improves upon previously manually-designed architectures. PMID:29690587

Data-driven automated acoustic analysis of human infant vocalizations using neural network tools.

PubMed

Warlaumont, Anne S; Oller, D Kimbrough; Buder, Eugene H; Dale, Rick; Kozma, Robert

2010-04-01

Acoustic analysis of infant vocalizations has typically employed traditional acoustic measures drawn from adult speech acoustics, such as f(0), duration, formant frequencies, amplitude, and pitch perturbation. Here an alternative and complementary method is proposed in which data-derived spectrographic features are central. 1-s-long spectrograms of vocalizations produced by six infants recorded longitudinally between ages 3 and 11 months are analyzed using a neural network consisting of a self-organizing map and a single-layer perceptron. The self-organizing map acquires a set of holistic, data-derived spectrographic receptive fields. The single-layer perceptron receives self-organizing map activations as input and is trained to classify utterances into prelinguistic phonatory categories (squeal, vocant, or growl), identify the ages at which they were produced, and identify the individuals who produced them. Classification performance was significantly better than chance for all three classification tasks. Performance is compared to another popular architecture, the fully supervised multilayer perceptron. In addition, the network's weights and patterns of activation are explored from several angles, for example, through traditional acoustic measurements of the network's receptive fields. Results support the use of this and related tools for deriving holistic acoustic features directly from infant vocalization data and for the automatic classification of infant vocalizations.

Neural Network Modeling for Gallium Arsenide IC Fabrication Process and Device Characteristics.

NASA Astrophysics Data System (ADS)

Creech, Gregory Lee, I.

This dissertation presents research focused on the utilization of neurocomputing technology to achieve enhanced yield and effective yield prediction in integrated circuit (IC) manufacturing. Artificial neural networks are employed to model complex relationships between material and device characteristics at critical stages of the semiconductor fabrication process. Whole wafer testing was performed on the starting substrate material and during wafer processing at four critical steps: Ohmic or Post-Contact, Post-Recess, Post-Gate and Final, i.e., at completion of fabrication. Measurements taken and subsequently used in modeling include, among others, doping concentrations, layer thicknesses, planar geometries, layer-to-layer alignments, resistivities, device voltages, and currents. The neural network architecture used in this research is the multilayer perceptron neural network (MLPNN). The MLPNN is trained in the supervised mode using the generalized delta learning rule. It has one hidden layer and uses continuous perceptrons. The research focuses on a number of different aspects. First is the development of inter-process stage models. Intermediate process stage models are created in a progressive fashion. Measurements of material and process/device characteristics taken at a specific processing stage and any previous stages are used as input to the model of the next processing stage characteristics. As the wafer moves through the fabrication process, measurements taken at all previous processing stages are used as input to each subsequent process stage model. Secondly, the development of neural network models for the estimation of IC parametric yield is demonstrated. Measurements of material and/or device characteristics taken at earlier fabrication stages are used to develop models of the final DC parameters. These characteristics are computed with the developed models and compared to acceptance windows to estimate the parametric yield. A sensitivity analysis is performed on the models developed during this yield estimation effort. This is accomplished by analyzing the total disturbance of network outputs due to perturbed inputs. When an input characteristic bears no, or little, statistical or deterministic relationship to the output characteristics, it can be removed as an input. Finally, neural network models are developed in the inverse direction. Characteristics measured after the final processing step are used as the input to model critical in-process characteristics. The modeled characteristics are used for whole wafer mapping and its statistical characterization. It is shown that this characterization can be accomplished with minimal in-process testing. The concepts and methodologies used in the development of the neural network models are presented. The modeling results are provided and compared to the actual measured values of each characteristic. An in-depth discussion of these results and ideas for future research are presented.

3D deeply supervised network for automated segmentation of volumetric medical images.

PubMed

Dou, Qi; Yu, Lequan; Chen, Hao; Jin, Yueming; Yang, Xin; Qin, Jing; Heng, Pheng-Ann

2017-10-01

While deep convolutional neural networks (CNNs) have achieved remarkable success in 2D medical image segmentation, it is still a difficult task for CNNs to segment important organs or structures from 3D medical images owing to several mutually affected challenges, including the complicated anatomical environments in volumetric images, optimization difficulties of 3D networks and inadequacy of training samples. In this paper, we present a novel and efficient 3D fully convolutional network equipped with a 3D deep supervision mechanism to comprehensively address these challenges; we call it 3D DSN. Our proposed 3D DSN is capable of conducting volume-to-volume learning and inference, which can eliminate redundant computations and alleviate the risk of over-fitting on limited training data. More importantly, the 3D deep supervision mechanism can effectively cope with the optimization problem of gradients vanishing or exploding when training a 3D deep model, accelerating the convergence speed and simultaneously improving the discrimination capability. Such a mechanism is developed by deriving an objective function that directly guides the training of both lower and upper layers in the network, so that the adverse effects of unstable gradient changes can be counteracted during the training procedure. We also employ a fully connected conditional random field model as a post-processing step to refine the segmentation results. We have extensively validated the proposed 3D DSN on two typical yet challenging volumetric medical image segmentation tasks: (i) liver segmentation from 3D CT scans and (ii) whole heart and great vessels segmentation from 3D MR images, by participating two grand challenges held in conjunction with MICCAI. We have achieved competitive segmentation results to state-of-the-art approaches in both challenges with a much faster speed, corroborating the effectiveness of our proposed 3D DSN. Copyright © 2017 Elsevier B.V. All rights reserved.

Automatic recognition of holistic functional brain networks using iteratively optimized convolutional neural networks (IO-CNN) with weak label initialization.

PubMed

Zhao, Yu; Ge, Fangfei; Liu, Tianming

2018-07-01

fMRI data decomposition techniques have advanced significantly from shallow models such as Independent Component Analysis (ICA) and Sparse Coding and Dictionary Learning (SCDL) to deep learning models such Deep Belief Networks (DBN) and Convolutional Autoencoder (DCAE). However, interpretations of those decomposed networks are still open questions due to the lack of functional brain atlases, no correspondence across decomposed or reconstructed networks across different subjects, and significant individual variabilities. Recent studies showed that deep learning, especially deep convolutional neural networks (CNN), has extraordinary ability of accommodating spatial object patterns, e.g., our recent works using 3D CNN for fMRI-derived network classifications achieved high accuracy with a remarkable tolerance for mistakenly labelled training brain networks. However, the training data preparation is one of the biggest obstacles in these supervised deep learning models for functional brain network map recognitions, since manual labelling requires tedious and time-consuming labours which will sometimes even introduce label mistakes. Especially for mapping functional networks in large scale datasets such as hundreds of thousands of brain networks used in this paper, the manual labelling method will become almost infeasible. In response, in this work, we tackled both the network recognition and training data labelling tasks by proposing a new iteratively optimized deep learning CNN (IO-CNN) framework with an automatic weak label initialization, which enables the functional brain networks recognition task to a fully automatic large-scale classification procedure. Our extensive experiments based on ABIDE-II 1099 brains' fMRI data showed the great promise of our IO-CNN framework. Copyright © 2018 Elsevier B.V. All rights reserved.

Invariant-feature-based adaptive automatic target recognition in obscured 3D point clouds

NASA Astrophysics Data System (ADS)

Khuon, Timothy; Kershner, Charles; Mattei, Enrico; Alverio, Arnel; Rand, Robert

2014-06-01

Target recognition and classification in a 3D point cloud is a non-trivial process due to the nature of the data collected from a sensor system. The signal can be corrupted by noise from the environment, electronic system, A/D converter, etc. Therefore, an adaptive system with a desired tolerance is required to perform classification and recognition optimally. The feature-based pattern recognition algorithm architecture as described below is particularly devised for solving a single-sensor classification non-parametrically. Feature set is extracted from an input point cloud, normalized, and classifier a neural network classifier. For instance, automatic target recognition in an urban area would require different feature sets from one in a dense foliage area. The figure above (see manuscript) illustrates the architecture of the feature based adaptive signature extraction of 3D point cloud including LIDAR, RADAR, and electro-optical data. This network takes a 3D cluster and classifies it into a specific class. The algorithm is a supervised and adaptive classifier with two modes: the training mode and the performing mode. For the training mode, a number of novel patterns are selected from actual or artificial data. A particular 3D cluster is input to the network as shown above for the decision class output. The network consists of three sequential functional modules. The first module is for feature extraction that extracts the input cluster into a set of singular value features or feature vector. Then the feature vector is input into the feature normalization module to normalize and balance it before being fed to the neural net classifier for the classification. The neural net can be trained by actual or artificial novel data until each trained output reaches the declared output within the defined tolerance. In case new novel data is added after the neural net has been learned, the training is then resumed until the neural net has incrementally learned with the new novel data. The associative memory capability of the neural net enables the incremental learning. The back propagation algorithm or support vector machine can be utilized for the classification and recognition.

Two-phase flow characterization based on advanced instrumentation, neural networks, and mathematical modeling

NASA Astrophysics Data System (ADS)

Mi, Ye

1998-12-01

The major objective of this thesis is focused on theoretical and experimental investigations of identifying and characterizing vertical and horizontal flow regimes in two-phase flows. A methodology of flow regime identification with impedance-based neural network systems and a comprehensive model of vertical slug flow have been developed. Vertical slug flow has been extensively investigated and characterized with geometric, kinematic and hydrodynamic parameters. A multi-sensor impedance void-meter and a multi-sensor magnetic flowmeter were developed. The impedance void-meter was cross-calibrated with other reliable techniques for void fraction measurements. The performance of the impedance void-meter to measure the void propagation velocity was evaluated by the drift flux model. It was proved that the magnetic flowmeter was applicable to vertical slug flow measurements. Separable signals from these instruments allow us to unearth most characteristics of vertical slug flow. A methodology of vertical flow regime identification was developed. Supervised neural network and self-organizing neural network systems were employed. First, they were trained with results from an idealized simulation of impedance in a two-phase mixture. The simulation was mainly based on Mishima and Ishii's flow regime map, the drift flux model, and the newly developed model of slug flow. Then, these trained systems were tested with impedance signals. The results showed that the neural network systems were appropriate classifiers of vertical flow regimes. The theoretical models and experimental databases used in the simulation were reliable. Furthermore, this approach was applied successfully to horizontal flow identification. A comprehensive model was developed to predict important characteristics of vertical slug flow. It was realized that the void fraction of the liquid slug is determined by the relative liquid motion between the Taylor bubble tail and the Taylor bubble wake. Relying on this understanding and experimental results, a special relationship was built for the void fraction of the liquid slug. The prediction of the void fraction of the liquid slug was considerably improved. Experimental characterization of vertical slug flows was performed extensively with the impedance void-meter and the magnetic flowmeter. The theoretical predictions were compared with the experimental results. The agreements between them are very satisfactory.

A novel end-to-end classifier using domain transferred deep convolutional neural networks for biomedical images.

PubMed

Pang, Shuchao; Yu, Zhezhou; Orgun, Mehmet A

2017-03-01

Highly accurate classification of biomedical images is an essential task in the clinical diagnosis of numerous medical diseases identified from those images. Traditional image classification methods combined with hand-crafted image feature descriptors and various classifiers are not able to effectively improve the accuracy rate and meet the high requirements of classification of biomedical images. The same also holds true for artificial neural network models directly trained with limited biomedical images used as training data or directly used as a black box to extract the deep features based on another distant dataset. In this study, we propose a highly reliable and accurate end-to-end classifier for all kinds of biomedical images via deep learning and transfer learning. We first apply domain transferred deep convolutional neural network for building a deep model; and then develop an overall deep learning architecture based on the raw pixels of original biomedical images using supervised training. In our model, we do not need the manual design of the feature space, seek an effective feature vector classifier or segment specific detection object and image patches, which are the main technological difficulties in the adoption of traditional image classification methods. Moreover, we do not need to be concerned with whether there are large training sets of annotated biomedical images, affordable parallel computing resources featuring GPUs or long times to wait for training a perfect deep model, which are the main problems to train deep neural networks for biomedical image classification as observed in recent works. With the utilization of a simple data augmentation method and fast convergence speed, our algorithm can achieve the best accuracy rate and outstanding classification ability for biomedical images. We have evaluated our classifier on several well-known public biomedical datasets and compared it with several state-of-the-art approaches. We propose a robust automated end-to-end classifier for biomedical images based on a domain transferred deep convolutional neural network model that shows a highly reliable and accurate performance which has been confirmed on several public biomedical image datasets. Copyright © 2017 Elsevier Ireland Ltd. All rights reserved.

Multilayer Extreme Learning Machine With Subnetwork Nodes for Representation Learning.

PubMed

Yang, Yimin; Wu, Q M Jonathan

2016-11-01

The extreme learning machine (ELM), which was originally proposed for "generalized" single-hidden layer feedforward neural networks, provides efficient unified learning solutions for the applications of clustering, regression, and classification. It presents competitive accuracy with superb efficiency in many applications. However, ELM with subnetwork nodes architecture has not attracted much research attentions. Recently, many methods have been proposed for supervised/unsupervised dimension reduction or representation learning, but these methods normally only work for one type of problem. This paper studies the general architecture of multilayer ELM (ML-ELM) with subnetwork nodes, showing that: 1) the proposed method provides a representation learning platform with unsupervised/supervised and compressed/sparse representation learning and 2) experimental results on ten image datasets and 16 classification datasets show that, compared to other conventional feature learning methods, the proposed ML-ELM with subnetwork nodes performs competitively or much better than other feature learning methods.

Quantifying the Evolution of Melt Ponds in the Marginal Ice Zone Using High Resolution Optical Imagery and Neural Networks

NASA Astrophysics Data System (ADS)

Ortiz, M.; Pinales, J. C.; Graber, H. C.; Wilkinson, J.; Lund, B.

2016-02-01

Melt ponds on sea ice play a significant and complex role on the thermodynamics in the Marginal Ice Zone (MIZ). Ponding reduces the sea ice's ability to reflect sunlight, and in consequence, exacerbates the albedo positive feedback cycle. In order to understand how melt ponds work and their effect on the heat uptake of sea ice, we must quantify ponds through their seasonal evolution first. A semi-supervised neural network three-class learning scheme using a gradient descent with momentum and adaptive learning rate backpropagation function is applied to classify melt ponds/melt areas in the Beaufort Sea region. The network uses high resolution panchromatic satellite images from the MEDEA program, which are collocated with autonomous platform arrays from the Marginal Ice Zone Program, including ice mass-balance buoys, arctic weather stations and wave buoys. The goal of the study is to capture the spatial variation of melt onset and freeze-up of the ponds within the MIZ, and gather ponding statistics such as size and concentration. The innovation of this work comes from training the neural network as the melt ponds evolve over time; making the machine learning algorithm time-dependent, which has not been previously done. We will achieve this by analyzing the image histograms through quantification of the minima and maxima intensity changes as well as linking textural variation information of the imagery. We will compare the evolution of the melt ponds against several different array sites on the sea ice to explore if there are spatial differences among the separated platforms in the MIZ.

Do deep convolutional neural networks really need to be deep when applied for remote scene classification?

NASA Astrophysics Data System (ADS)

Luo, Chang; Wang, Jie; Feng, Gang; Xu, Suhui; Wang, Shiqiang

2017-10-01

Deep convolutional neural networks (CNNs) have been widely used to obtain high-level representation in various computer vision tasks. However, for remote scene classification, there are not sufficient images to train a very deep CNN from scratch. From two viewpoints of generalization power, we propose two promising kinds of deep CNNs for remote scenes and try to find whether deep CNNs need to be deep for remote scene classification. First, we transfer successful pretrained deep CNNs to remote scenes based on the theory that depth of CNNs brings the generalization power by learning available hypothesis for finite data samples. Second, according to the opposite viewpoint that generalization power of deep CNNs comes from massive memorization and shallow CNNs with enough neural nodes have perfect finite sample expressivity, we design a lightweight deep CNN (LDCNN) for remote scene classification. With five well-known pretrained deep CNNs, experimental results on two independent remote-sensing datasets demonstrate that transferred deep CNNs can achieve state-of-the-art results in an unsupervised setting. However, because of its shallow architecture, LDCNN cannot obtain satisfactory performance, regardless of whether in an unsupervised, semisupervised, or supervised setting. CNNs really need depth to obtain general features for remote scenes. This paper also provides baseline for applying deep CNNs to other remote sensing tasks.

Artificial Neural Identification and LMI Transformation for Model Reduction-Based Control of the Buck Switch-Mode Regulator

NASA Astrophysics Data System (ADS)

Al-Rabadi, Anas N.

2009-10-01

This research introduces a new method of intelligent control for the control of the Buck converter using newly developed small signal model of the pulse width modulation (PWM) switch. The new method uses supervised neural network to estimate certain parameters of the transformed system matrix [Ã]. Then, a numerical algorithm used in robust control called linear matrix inequality (LMI) optimization technique is used to determine the permutation matrix [P] so that a complete system transformation {[B˜], [C˜], [Ẽ]} is possible. The transformed model is then reduced using the method of singular perturbation, and state feedback control is applied to enhance system performance. The experimental results show that the new control methodology simplifies the model in the Buck converter and thus uses a simpler controller that produces the desired system response for performance enhancement.

Automated diagnosis of rolling bearings using MRA and neural networks

NASA Astrophysics Data System (ADS)

Castejón, C.; Lara, O.; García-Prada, J. C.

2010-01-01

Any industry needs an efficient predictive plan in order to optimize the management of resources and improve the economy of the plant by reducing unnecessary costs and increasing the level of safety. A great percentage of breakdowns in productive processes are caused by bearings. They begin to deteriorate from early stages of their functional life, also called the incipient level. This manuscript develops an automated diagnosis of rolling bearings based on the analysis and classification of signature vibrations. The novelty of this work is the application of the methodology proposed for data collected from a quasi-real industrial machine, where rolling bearings support the radial and axial loads the bearings are designed for. Multiresolution analysis (MRA) is used in a first stage in order to extract the most interesting features from signals. Features will be used in a second stage as inputs of a supervised neural network (NN) for classification purposes. Experimental results carried out in a real system show the soundness of the method which detects four bearing conditions (normal, inner race fault, outer race fault and ball fault) in a very incipient stage.

Entity recognition in the biomedical domain using a hybrid approach.

PubMed

Basaldella, Marco; Furrer, Lenz; Tasso, Carlo; Rinaldi, Fabio

2017-11-09

This article describes a high-recall, high-precision approach for the extraction of biomedical entities from scientific articles. The approach uses a two-stage pipeline, combining a dictionary-based entity recognizer with a machine-learning classifier. First, the OGER entity recognizer, which has a bias towards high recall, annotates the terms that appear in selected domain ontologies. Subsequently, the Distiller framework uses this information as a feature for a machine learning algorithm to select the relevant entities only. For this step, we compare two different supervised machine-learning algorithms: Conditional Random Fields and Neural Networks. In an in-domain evaluation using the CRAFT corpus, we test the performance of the combined systems when recognizing chemicals, cell types, cellular components, biological processes, molecular functions, organisms, proteins, and biological sequences. Our best system combines dictionary-based candidate generation with Neural-Network-based filtering. It achieves an overall precision of 86% at a recall of 60% on the named entity recognition task, and a precision of 51% at a recall of 49% on the concept recognition task. These results are to our knowledge the best reported so far in this particular task.

Emotional recognition from the speech signal for a virtual education agent

NASA Astrophysics Data System (ADS)

Tickle, A.; Raghu, S.; Elshaw, M.

2013-06-01

This paper explores the extraction of features from the speech wave to perform intelligent emotion recognition. A feature extract tool (openSmile) was used to obtain a baseline set of 998 acoustic features from a set of emotional speech recordings from a microphone. The initial features were reduced to the most important ones so recognition of emotions using a supervised neural network could be performed. Given that the future use of virtual education agents lies with making the agents more interactive, developing agents with the capability to recognise and adapt to the emotional state of humans is an important step.

Fabric defect detection based on visual saliency using deep feature and low-rank recovery

NASA Astrophysics Data System (ADS)

Liu, Zhoufeng; Wang, Baorui; Li, Chunlei; Li, Bicao; Dong, Yan

2018-04-01

Fabric defect detection plays an important role in improving the quality of fabric product. In this paper, a novel fabric defect detection method based on visual saliency using deep feature and low-rank recovery was proposed. First, unsupervised training is carried out by the initial network parameters based on MNIST large datasets. The supervised fine-tuning of fabric image library based on Convolutional Neural Networks (CNNs) is implemented, and then more accurate deep neural network model is generated. Second, the fabric images are uniformly divided into the image block with the same size, then we extract their multi-layer deep features using the trained deep network. Thereafter, all the extracted features are concentrated into a feature matrix. Third, low-rank matrix recovery is adopted to divide the feature matrix into the low-rank matrix which indicates the background and the sparse matrix which indicates the salient defect. In the end, the iterative optimal threshold segmentation algorithm is utilized to segment the saliency maps generated by the sparse matrix to locate the fabric defect area. Experimental results demonstrate that the feature extracted by CNN is more suitable for characterizing the fabric texture than the traditional LBP, HOG and other hand-crafted features extraction method, and the proposed method can accurately detect the defect regions of various fabric defects, even for the image with complex texture.

Performance evaluation of MLP and RBF feed forward neural network for the recognition of off-line handwritten characters

NASA Astrophysics Data System (ADS)

Rishi, Rahul; Choudhary, Amit; Singh, Ravinder; Dhaka, Vijaypal Singh; Ahlawat, Savita; Rao, Mukta

2010-02-01

In this paper we propose a system for classification problem of handwritten text. The system is composed of preprocessing module, supervised learning module and recognition module on a very broad level. The preprocessing module digitizes the documents and extracts features (tangent values) for each character. The radial basis function network is used in the learning and recognition modules. The objective is to analyze and improve the performance of Multi Layer Perceptron (MLP) using RBF transfer functions over Logarithmic Sigmoid Function. The results of 35 experiments indicate that the Feed Forward MLP performs accurately and exhaustively with RBF. With the change in weight update mechanism and feature-drawn preprocessing module, the proposed system is competent with good recognition show.

Human Splice-Site Prediction with Deep Neural Networks.

PubMed

Naito, Tatsuhiko

2018-04-18

Accurate splice-site prediction is essential to delineate gene structures from sequence data. Several computational techniques have been applied to create a system to predict canonical splice sites. For classification tasks, deep neural networks (DNNs) have achieved record-breaking results and often outperformed other supervised learning techniques. In this study, a new method of splice-site prediction using DNNs was proposed. The proposed system receives an input sequence data and returns an answer as to whether it is splice site. The length of input is 140 nucleotides, with the consensus sequence (i.e., "GT" and "AG" for the donor and acceptor sites, respectively) in the middle. Each input sequence model is applied to the pretrained DNN model that determines the probability that an input is a splice site. The model consists of convolutional layers and bidirectional long short-term memory network layers. The pretraining and validation were conducted using the data set tested in previously reported methods. The performance evaluation results showed that the proposed method can outperform the previous methods. In addition, the pattern learned by the DNNs was visualized as position frequency matrices (PFMs). Some of PFMs were very similar to the consensus sequence. The trained DNN model and the brief source code for the prediction system are uploaded. Further improvement will be achieved following the further development of DNNs.

Distant supervision for neural relation extraction integrated with word attention and property features.

PubMed

Qu, Jianfeng; Ouyang, Dantong; Hua, Wen; Ye, Yuxin; Li, Ximing

2018-04-01

Distant supervision for neural relation extraction is an efficient approach to extracting massive relations with reference to plain texts. However, the existing neural methods fail to capture the critical words in sentence encoding and meanwhile lack useful sentence information for some positive training instances. To address the above issues, we propose a novel neural relation extraction model. First, we develop a word-level attention mechanism to distinguish the importance of each individual word in a sentence, increasing the attention weights for those critical words. Second, we investigate the semantic information from word embeddings of target entities, which can be developed as a supplementary feature for the extractor. Experimental results show that our model outperforms previous state-of-the-art baselines. Copyright © 2018 Elsevier Ltd. All rights reserved.

Predicting non-linear dynamics by stable local learning in a recurrent spiking neural network.

PubMed

Gilra, Aditya; Gerstner, Wulfram

2017-11-27

The brain needs to predict how the body reacts to motor commands, but how a network of spiking neurons can learn non-linear body dynamics using local, online and stable learning rules is unclear. Here, we present a supervised learning scheme for the feedforward and recurrent connections in a network of heterogeneous spiking neurons. The error in the output is fed back through fixed random connections with a negative gain, causing the network to follow the desired dynamics. The rule for Feedback-based Online Local Learning Of Weights (FOLLOW) is local in the sense that weight changes depend on the presynaptic activity and the error signal projected onto the postsynaptic neuron. We provide examples of learning linear, non-linear and chaotic dynamics, as well as the dynamics of a two-link arm. Under reasonable approximations, we show, using the Lyapunov method, that FOLLOW learning is uniformly stable, with the error going to zero asymptotically.

Predicting non-linear dynamics by stable local learning in a recurrent spiking neural network

PubMed Central

Gerstner, Wulfram

2017-01-01

The brain needs to predict how the body reacts to motor commands, but how a network of spiking neurons can learn non-linear body dynamics using local, online and stable learning rules is unclear. Here, we present a supervised learning scheme for the feedforward and recurrent connections in a network of heterogeneous spiking neurons. The error in the output is fed back through fixed random connections with a negative gain, causing the network to follow the desired dynamics. The rule for Feedback-based Online Local Learning Of Weights (FOLLOW) is local in the sense that weight changes depend on the presynaptic activity and the error signal projected onto the postsynaptic neuron. We provide examples of learning linear, non-linear and chaotic dynamics, as well as the dynamics of a two-link arm. Under reasonable approximations, we show, using the Lyapunov method, that FOLLOW learning is uniformly stable, with the error going to zero asymptotically. PMID:29173280

Spiking neuron network Helmholtz machine.

PubMed

Sountsov, Pavel; Miller, Paul

2015-01-01

An increasing amount of behavioral and neurophysiological data suggests that the brain performs optimal (or near-optimal) probabilistic inference and learning during perception and other tasks. Although many machine learning algorithms exist that perform inference and learning in an optimal way, the complete description of how one of those algorithms (or a novel algorithm) can be implemented in the brain is currently incomplete. There have been many proposed solutions that address how neurons can perform optimal inference but the question of how synaptic plasticity can implement optimal learning is rarely addressed. This paper aims to unify the two fields of probabilistic inference and synaptic plasticity by using a neuronal network of realistic model spiking neurons to implement a well-studied computational model called the Helmholtz Machine. The Helmholtz Machine is amenable to neural implementation as the algorithm it uses to learn its parameters, called the wake-sleep algorithm, uses a local delta learning rule. Our spiking-neuron network implements both the delta rule and a small example of a Helmholtz machine. This neuronal network can learn an internal model of continuous-valued training data sets without supervision. The network can also perform inference on the learned internal models. We show how various biophysical features of the neural implementation constrain the parameters of the wake-sleep algorithm, such as the duration of the wake and sleep phases of learning and the minimal sample duration. We examine the deviations from optimal performance and tie them to the properties of the synaptic plasticity rule.

Spiking neuron network Helmholtz machine

PubMed Central

Sountsov, Pavel; Miller, Paul

2015-01-01

An increasing amount of behavioral and neurophysiological data suggests that the brain performs optimal (or near-optimal) probabilistic inference and learning during perception and other tasks. Although many machine learning algorithms exist that perform inference and learning in an optimal way, the complete description of how one of those algorithms (or a novel algorithm) can be implemented in the brain is currently incomplete. There have been many proposed solutions that address how neurons can perform optimal inference but the question of how synaptic plasticity can implement optimal learning is rarely addressed. This paper aims to unify the two fields of probabilistic inference and synaptic plasticity by using a neuronal network of realistic model spiking neurons to implement a well-studied computational model called the Helmholtz Machine. The Helmholtz Machine is amenable to neural implementation as the algorithm it uses to learn its parameters, called the wake-sleep algorithm, uses a local delta learning rule. Our spiking-neuron network implements both the delta rule and a small example of a Helmholtz machine. This neuronal network can learn an internal model of continuous-valued training data sets without supervision. The network can also perform inference on the learned internal models. We show how various biophysical features of the neural implementation constrain the parameters of the wake-sleep algorithm, such as the duration of the wake and sleep phases of learning and the minimal sample duration. We examine the deviations from optimal performance and tie them to the properties of the synaptic plasticity rule. PMID:25954191

Evaluation of Deep Learning Representations of Spatial Storm Data

NASA Astrophysics Data System (ADS)

Gagne, D. J., II; Haupt, S. E.; Nychka, D. W.

2017-12-01

The spatial structure of a severe thunderstorm and its surrounding environment provide useful information about the potential for severe weather hazards, including tornadoes, hail, and high winds. Statistics computed over the area of a storm or from the pre-storm environment can provide descriptive information but fail to capture structural information. Because the storm environment is a complex, high-dimensional space, identifying methods to encode important spatial storm information in a low-dimensional form should aid analysis and prediction of storms by statistical and machine learning models. Principal component analysis (PCA), a more traditional approach, transforms high-dimensional data into a set of linearly uncorrelated, orthogonal components ordered by the amount of variance explained by each component. The burgeoning field of deep learning offers two potential approaches to this problem. Convolutional Neural Networks are a supervised learning method for transforming spatial data into a hierarchical set of feature maps that correspond with relevant combinations of spatial structures in the data. Generative Adversarial Networks (GANs) are an unsupervised deep learning model that uses two neural networks trained against each other to produce encoded representations of spatial data. These different spatial encoding methods were evaluated on the prediction of severe hail for a large set of storm patches extracted from the NCAR convection-allowing ensemble. Each storm patch contains information about storm structure and the near-storm environment. Logistic regression and random forest models were trained using the PCA and GAN encodings of the storm data and were compared against the predictions from a convolutional neural network. All methods showed skill over climatology at predicting the probability of severe hail. However, the verification scores among the methods were very similar and the predictions were highly correlated. Further evaluations are being performed to determine how the choice of input variables affects the results.

An immune-inspired semi-supervised algorithm for breast cancer diagnosis.

PubMed

Peng, Lingxi; Chen, Wenbin; Zhou, Wubai; Li, Fufang; Yang, Jin; Zhang, Jiandong

2016-10-01

Breast cancer is the most frequently and world widely diagnosed life-threatening cancer, which is the leading cause of cancer death among women. Early accurate diagnosis can be a big plus in treating breast cancer. Researchers have approached this problem using various data mining and machine learning techniques such as support vector machine, artificial neural network, etc. The computer immunology is also an intelligent method inspired by biological immune system, which has been successfully applied in pattern recognition, combination optimization, machine learning, etc. However, most of these diagnosis methods belong to a supervised diagnosis method. It is very expensive to obtain labeled data in biology and medicine. In this paper, we seamlessly integrate the state-of-the-art research on life science with artificial intelligence, and propose a semi-supervised learning algorithm to reduce the need for labeled data. We use two well-known benchmark breast cancer datasets in our study, which are acquired from the UCI machine learning repository. Extensive experiments are conducted and evaluated on those two datasets. Our experimental results demonstrate the effectiveness and efficiency of our proposed algorithm, which proves that our algorithm is a promising automatic diagnosis method for breast cancer. Copyright © 2016 Elsevier Ireland Ltd. All rights reserved.

Radiation dose reduction in digital breast tomosynthesis (DBT) by means of deep-learning-based supervised image processing

NASA Astrophysics Data System (ADS)

Liu, Junchi; Zarshenas, Amin; Qadir, Ammar; Wei, Zheng; Yang, Limin; Fajardo, Laurie; Suzuki, Kenji

2018-03-01

To reduce cumulative radiation exposure and lifetime risks for radiation-induced cancer from breast cancer screening, we developed a deep-learning-based supervised image-processing technique called neural network convolution (NNC) for radiation dose reduction in DBT. NNC employed patched-based neural network regression in a convolutional manner to convert lower-dose (LD) to higher-dose (HD) tomosynthesis images. We trained our NNC with quarter-dose (25% of the standard dose: 12 mAs at 32 kVp) raw projection images and corresponding "teaching" higher-dose (HD) images (200% of the standard dose: 99 mAs at 32 kVp) of a breast cadaver phantom acquired with a DBT system (Selenia Dimensions, Hologic, CA). Once trained, NNC no longer requires HD images. It converts new LD images to images that look like HD images; thus the term "virtual" HD (VHD) images. We reconstructed tomosynthesis slices on a research DBT system. To determine a dose reduction rate, we acquired 4 studies of another test phantom at 4 different radiation doses (1.35, 2.7, 4.04, and 5.39 mGy entrance dose). Structural SIMilarity (SSIM) index was used to evaluate the image quality. For testing, we collected half-dose (50% of the standard dose: 32+/-14 mAs at 33+/-5 kVp) and full-dose (standard dose: 68+/-23 mAs at 33+/-5 kvp) images of 10 clinical cases with the DBT system at University of Iowa Hospitals and Clinics. NNC converted half-dose DBT images of 10 clinical cases to VHD DBT images that were equivalent to full dose DBT images. Our cadaver phantom experiment demonstrated 79% dose reduction.

Supervised learning methods for pathological arterial pulse wave differentiation: A SVM and neural networks approach.

PubMed

Paiva, Joana S; Cardoso, João; Pereira, Tânia

2018-01-01

The main goal of this study was to develop an automatic method based on supervised learning methods, able to distinguish healthy from pathologic arterial pulse wave (APW), and those two from noisy waveforms (non-relevant segments of the signal), from the data acquired during a clinical examination with a novel optical system. The APW dataset analysed was composed by signals acquired in a clinical environment from a total of 213 subjects, including healthy volunteers and non-healthy patients. The signals were parameterised by means of 39pulse features: morphologic, time domain statistics, cross-correlation features, wavelet features. Multiclass Support Vector Machine Recursive Feature Elimination (SVM RFE) method was used to select the most relevant features. A comparative study was performed in order to evaluate the performance of the two classifiers: Support Vector Machine (SVM) and Artificial Neural Network (ANN). SVM achieved a statistically significant better performance for this problem with an average accuracy of 0.9917±0.0024 and a F-Measure of 0.9925±0.0019, in comparison with ANN, which reached the values of 0.9847±0.0032 and 0.9852±0.0031 for Accuracy and F-Measure, respectively. A significant difference was observed between the performances obtained with SVM classifier using a different number of features from the original set available. The comparison between SVM and NN allowed reassert the higher performance of SVM. The results obtained in this study showed the potential of the proposed method to differentiate those three important signal outcomes (healthy, pathologic and noise) and to reduce bias associated with clinical diagnosis of cardiovascular disease using APW. Copyright © 2017 Elsevier B.V. All rights reserved.

Mapping of rock types using a joint approach by combining the multivariate statistics, self-organizing map and Bayesian neural networks: an example from IODP 323 site

NASA Astrophysics Data System (ADS)

Karmakar, Mampi; Maiti, Saumen; Singh, Amrita; Ojha, Maheswar; Maity, Bhabani Sankar

2017-07-01

Modeling and classification of the subsurface lithology is very important to understand the evolution of the earth system. However, precise classification and mapping of lithology using a single framework are difficult due to the complexity and the nonlinearity of the problem driven by limited core sample information. Here, we implement a joint approach by combining the unsupervised and the supervised methods in a single framework for better classification and mapping of rock types. In the unsupervised method, we use the principal component analysis (PCA), K-means cluster analysis (K-means), dendrogram analysis, Fuzzy C-means (FCM) cluster analysis and self-organizing map (SOM). In the supervised method, we use the Bayesian neural networks (BNN) optimized by the Hybrid Monte Carlo (HMC) (BNN-HMC) and the scaled conjugate gradient (SCG) (BNN-SCG) techniques. We use P-wave velocity, density, neutron porosity, resistivity and gamma ray logs of the well U1343E of the Integrated Ocean Drilling Program (IODP) Expedition 323 in the Bering Sea slope region. While the SOM algorithm allows us to visualize the clustering results in spatial domain, the combined classification schemes (supervised and unsupervised) uncover the different patterns of lithology such of as clayey-silt, diatom-silt and silty-clay from an un-cored section of the drilled hole. In addition, the BNN approach is capable of estimating uncertainty in the predictive modeling of three types of rocks over the entire lithology section at site U1343. Alternate succession of clayey-silt, diatom-silt and silty-clay may be representative of crustal inhomogeneity in general and thus could be a basis for detail study related to the productivity of methane gas in the oceans worldwide. Moreover, at the 530 m depth down below seafloor (DSF), the transition from Pliocene to Pleistocene could be linked to lithological alternation between the clayey-silt and the diatom-silt. The present results could provide the basis for the detailed study to get deeper insight into the Bering Sea' sediment deposition and sequence.

The application of artificial neural networks in astronomy

NASA Astrophysics Data System (ADS)

Li, Li-Li; Zhang, Yan-Xia; Zhao, Yong-Heng; Yang, Da-Wei

2006-12-01

Artificial Neural Networks (ANNs) are computer algorithms inspired from simple models of human central nervous system activity. They can be roughly divided into two main kinds: supervised and unsupervised. The supervised approach lays the stress on "teaching" a machine to do the work of a mention human expert, usually by showing examples for which the true answer is supplied by the expert. The unsupervised one is aimed at learning new things from the data, and most useful when the data cannot easily be plotted in a two or three dimensional space. ANNs have been used widely and successfully in various fields, for instance, pattern recognition, financial analysis, biology, engineering and so on, because they have many merits such as self-learning, self-adapting, good robustness and dynamically rapid response as well as strong capability of dealing with non-linear problems. In the last few years there has been an increasing interest toward the astronomical applications of ANNs. In this paper, the authors firstly introduce the fundamental principle of ANNs together with the architecture of the network and outline various kinds of learning algorithms and network toplogies. The specific aspects of the applications of ANNs in astronomical problems are also listed, which contain the strong capabilities of approximating to arbitrary accuracy, any nonlinear functional mapping, parallel and distributed storage, tolerance of faulty and generalization of results. They summarize the advantages and disadvantages of main ANN models available to the astronomical community. Furthermore, the application cases of ANNs in astronomy are mainly described in detail. Here, the focus is on some of the most interesting fields of its application, for example: object detection, star/galaxy classification, spectral classification, galaxy morphology classification, the estimation of photometric redshifts of galaxies and time series analysis. In addition, other kinds of applications have been only touched upon. Finally, the development and application prospects of ANNs is discussed. With the increase of quantity and the distributing complexity of astronomical data, its scientific exploitation requires a variety of automated tools, which are capable to perform huge amount of work, such as data preprocessing, feature selection, data reduction, data mining amd data analysis. ANNs, one of intelligent tools, will show more and more superiorities.

Unsupervised learning of digit recognition using spike-timing-dependent plasticity

PubMed Central

Diehl, Peter U.; Cook, Matthew

2015-01-01

In order to understand how the mammalian neocortex is performing computations, two things are necessary; we need to have a good understanding of the available neuronal processing units and mechanisms, and we need to gain a better understanding of how those mechanisms are combined to build functioning systems. Therefore, in recent years there is an increasing interest in how spiking neural networks (SNN) can be used to perform complex computations or solve pattern recognition tasks. However, it remains a challenging task to design SNNs which use biologically plausible mechanisms (especially for learning new patterns), since most such SNN architectures rely on training in a rate-based network and subsequent conversion to a SNN. We present a SNN for digit recognition which is based on mechanisms with increased biological plausibility, i.e., conductance-based instead of current-based synapses, spike-timing-dependent plasticity with time-dependent weight change, lateral inhibition, and an adaptive spiking threshold. Unlike most other systems, we do not use a teaching signal and do not present any class labels to the network. Using this unsupervised learning scheme, our architecture achieves 95% accuracy on the MNIST benchmark, which is better than previous SNN implementations without supervision. The fact that we used no domain-specific knowledge points toward the general applicability of our network design. Also, the performance of our network scales well with the number of neurons used and shows similar performance for four different learning rules, indicating robustness of the full combination of mechanisms, which suggests applicability in heterogeneous biological neural networks. PMID:26941637

Load reduction test method of similarity theory and BP neural networks of large cranes

NASA Astrophysics Data System (ADS)

Yang, Ruigang; Duan, Zhibin; Lu, Yi; Wang, Lei; Xu, Gening

2016-01-01

Static load tests are an important means of supervising and detecting a crane's lift capacity. Due to space restrictions, however, there are difficulties and potential danger when testing large bridge cranes. To solve the loading problems of large-tonnage cranes during testing, an equivalency test is proposed based on the similarity theory and BP neural networks. The maximum stress and displacement of a large bridge crane is tested in small loads, combined with the training neural network of a similar structure crane through stress and displacement data which is collected by a physics simulation progressively loaded to a static load test load within the material scope of work. The maximum stress and displacement of a crane under a static load test load can be predicted through the relationship of stress, displacement, and load. By measuring the stress and displacement of small tonnage weights, the stress and displacement of large loads can be predicted, such as the maximum load capacity, which is 1.25 times the rated capacity. Experimental study shows that the load reduction test method can reflect the lift capacity of large bridge cranes. The load shedding predictive analysis for Sanxia 1200 t bridge crane test data indicates that when the load is 1.25 times the rated lifting capacity, the predicted displacement and actual displacement error is zero. The method solves the problem that lifting capacities are difficult to obtain and testing accidents are easily possible when 1.25 times related weight loads are tested for large tonnage cranes.

A computational framework for the detection of subcortical brain dysmaturation in neonatal MRI using 3D Convolutional Neural Networks.

PubMed

Ceschin, Rafael; Zahner, Alexandria; Reynolds, William; Gaesser, Jenna; Zuccoli, Giulio; Lo, Cecilia W; Gopalakrishnan, Vanathi; Panigrahy, Ashok

2018-05-21

Deep neural networks are increasingly being used in both supervised learning for classification tasks and unsupervised learning to derive complex patterns from the input data. However, the successful implementation of deep neural networks using neuroimaging datasets requires adequate sample size for training and well-defined signal intensity based structural differentiation. There is a lack of effective automated diagnostic tools for the reliable detection of brain dysmaturation in the neonatal period, related to small sample size and complex undifferentiated brain structures, despite both translational research and clinical importance. Volumetric information alone is insufficient for diagnosis. In this study, we developed a computational framework for the automated classification of brain dysmaturation from neonatal MRI, by combining a specific deep neural network implementation with neonatal structural brain segmentation as a method for both clinical pattern recognition and data-driven inference into the underlying structural morphology. We implemented three-dimensional convolution neural networks (3D-CNNs) to specifically classify dysplastic cerebelli, a subset of surface-based subcortical brain dysmaturation, in term infants born with congenital heart disease. We obtained a 0.985 ± 0. 0241-classification accuracy of subtle cerebellar dysplasia in CHD using 10-fold cross-validation. Furthermore, the hidden layer activations and class activation maps depicted regional vulnerability of the superior surface of the cerebellum, (composed of mostly the posterior lobe and the midline vermis), in regards to differentiating the dysplastic process from normal tissue. The posterior lobe and the midline vermis provide regional differentiation that is relevant to not only to the clinical diagnosis of cerebellar dysplasia, but also genetic mechanisms and neurodevelopmental outcome correlates. These findings not only contribute to the detection and classification of a subset of neonatal brain dysmaturation, but also provide insight to the pathogenesis of cerebellar dysplasia in CHD. In addition, this is one of the first examples of the application of deep learning to a neuroimaging dataset, in which the hidden layer activation revealed diagnostically and biologically relevant features about the clinical pathogenesis. The code developed for this project is open source, published under the BSD License, and designed to be generalizable to applications both within and beyond neonatal brain imaging. Copyright © 2018 Elsevier Inc. All rights reserved.

QUEST: Eliminating Online Supervised Learning for Efficient Classification Algorithms.

PubMed

Zwartjes, Ardjan; Havinga, Paul J M; Smit, Gerard J M; Hurink, Johann L

2016-10-01

In this work, we introduce QUEST (QUantile Estimation after Supervised Training), an adaptive classification algorithm for Wireless Sensor Networks (WSNs) that eliminates the necessity for online supervised learning. Online processing is important for many sensor network applications. Transmitting raw sensor data puts high demands on the battery, reducing network life time. By merely transmitting partial results or classifications based on the sampled data, the amount of traffic on the network can be significantly reduced. Such classifications can be made by learning based algorithms using sampled data. An important issue, however, is the training phase of these learning based algorithms. Training a deployed sensor network requires a lot of communication and an impractical amount of human involvement. QUEST is a hybrid algorithm that combines supervised learning in a controlled environment with unsupervised learning on the location of deployment. Using the SITEX02 dataset, we demonstrate that the presented solution works with a performance penalty of less than 10% in 90% of the tests. Under some circumstances, it even outperforms a network of classifiers completely trained with supervised learning. As a result, the need for on-site supervised learning and communication for training is completely eliminated by our solution.

Comparative Analysis of River Flow Modelling by Using Supervised Learning Technique

NASA Astrophysics Data System (ADS)

Ismail, Shuhaida; Mohamad Pandiahi, Siraj; Shabri, Ani; Mustapha, Aida

2018-04-01

The goal of this research is to investigate the efficiency of three supervised learning algorithms for forecasting monthly river flow of the Indus River in Pakistan, spread over 550 square miles or 1800 square kilometres. The algorithms include the Least Square Support Vector Machine (LSSVM), Artificial Neural Network (ANN) and Wavelet Regression (WR). The forecasting models predict the monthly river flow obtained from the three models individually for river flow data and the accuracy of the all models were then compared against each other. The monthly river flow of the said river has been forecasted using these three models. The obtained results were compared and statistically analysed. Then, the results of this analytical comparison showed that LSSVM model is more precise in the monthly river flow forecasting. It was found that LSSVM has he higher r with the value of 0.934 compared to other models. This indicate that LSSVM is more accurate and efficient as compared to the ANN and WR model.

Semi-supervised vibration-based classification and condition monitoring of compressors

NASA Astrophysics Data System (ADS)

Potočnik, Primož; Govekar, Edvard

2017-09-01

Semi-supervised vibration-based classification and condition monitoring of the reciprocating compressors installed in refrigeration appliances is proposed in this paper. The method addresses the problem of industrial condition monitoring where prior class definitions are often not available or difficult to obtain from local experts. The proposed method combines feature extraction, principal component analysis, and statistical analysis for the extraction of initial class representatives, and compares the capability of various classification methods, including discriminant analysis (DA), neural networks (NN), support vector machines (SVM), and extreme learning machines (ELM). The use of the method is demonstrated on a case study which was based on industrially acquired vibration measurements of reciprocating compressors during the production of refrigeration appliances. The paper presents a comparative qualitative analysis of the applied classifiers, confirming the good performance of several nonlinear classifiers. If the model parameters are properly selected, then very good classification performance can be obtained from NN trained by Bayesian regularization, SVM and ELM classifiers. The method can be effectively applied for the industrial condition monitoring of compressors.

Unsupervised, Robust Estimation-based Clustering for Multispectral Images

NASA Technical Reports Server (NTRS)

Netanyahu, Nathan S.

1997-01-01

To prepare for the challenge of handling the archiving and querying of terabyte-sized scientific spatial databases, the NASA Goddard Space Flight Center's Applied Information Sciences Branch (AISB, Code 935) developed a number of characterization algorithms that rely on supervised clustering techniques. The research reported upon here has been aimed at continuing the evolution of some of these supervised techniques, namely the neural network and decision tree-based classifiers, plus extending the approach to incorporating unsupervised clustering algorithms, such as those based on robust estimation (RE) techniques. The algorithms developed under this task should be suited for use by the Intelligent Information Fusion System (IIFS) metadata extraction modules, and as such these algorithms must be fast, robust, and anytime in nature. Finally, so that the planner/schedule module of the IlFS can oversee the use and execution of these algorithms, all information required by the planner/scheduler must be provided to the IIFS development team to ensure the timely integration of these algorithms into the overall system.

Using virtual data for training deep model for hand gesture recognition

NASA Astrophysics Data System (ADS)

Nikolaev, E. I.; Dvoryaninov, P. V.; Lensky, Y. Y.; Drozdovsky, N. S.

2018-05-01

Deep learning has shown real promise for the classification efficiency for hand gesture recognition problems. In this paper, the authors present experimental results for a deeply-trained model for hand gesture recognition through the use of hand images. The authors have trained two deep convolutional neural networks. The first architecture produces the hand position as a 2D-vector by input hand image. The second one predicts the hand gesture class for the input image. The first proposed architecture produces state of the art results with an accuracy rate of 89% and the second architecture with split input produces accuracy rate of 85.2%. In this paper, the authors also propose using virtual data for training a supervised deep model. Such technique is aimed to avoid using original labelled images in the training process. The interest of this method in data preparation is motivated by the need to overcome one of the main challenges of deep supervised learning: using a copious amount of labelled data during training.

DL-ReSuMe: A Delay Learning-Based Remote Supervised Method for Spiking Neurons.

PubMed

Taherkhani, Aboozar; Belatreche, Ammar; Li, Yuhua; Maguire, Liam P

2015-12-01

Recent research has shown the potential capability of spiking neural networks (SNNs) to model complex information processing in the brain. There is biological evidence to prove the use of the precise timing of spikes for information coding. However, the exact learning mechanism in which the neuron is trained to fire at precise times remains an open problem. The majority of the existing learning methods for SNNs are based on weight adjustment. However, there is also biological evidence that the synaptic delay is not constant. In this paper, a learning method for spiking neurons, called delay learning remote supervised method (DL-ReSuMe), is proposed to merge the delay shift approach and ReSuMe-based weight adjustment to enhance the learning performance. DL-ReSuMe uses more biologically plausible properties, such as delay learning, and needs less weight adjustment than ReSuMe. Simulation results have shown that the proposed DL-ReSuMe approach achieves learning accuracy and learning speed improvements compared with ReSuMe.

Long short-term memory for speaker generalization in supervised speech separation

PubMed Central

Chen, Jitong; Wang, DeLiang

2017-01-01

Speech separation can be formulated as learning to estimate a time-frequency mask from acoustic features extracted from noisy speech. For supervised speech separation, generalization to unseen noises and unseen speakers is a critical issue. Although deep neural networks (DNNs) have been successful in noise-independent speech separation, DNNs are limited in modeling a large number of speakers. To improve speaker generalization, a separation model based on long short-term memory (LSTM) is proposed, which naturally accounts for temporal dynamics of speech. Systematic evaluation shows that the proposed model substantially outperforms a DNN-based model on unseen speakers and unseen noises in terms of objective speech intelligibility. Analyzing LSTM internal representations reveals that LSTM captures long-term speech contexts. It is also found that the LSTM model is more advantageous for low-latency speech separation and it, without future frames, performs better than the DNN model with future frames. The proposed model represents an effective approach for speaker- and noise-independent speech separation. PMID:28679261

Facebook for Supervision? Research Education Shaped by the Structural Properties of a Social Media Space

ERIC Educational Resources Information Center

Pimmer, Christoph; Chipps, Jennifer; Brysiewicz, Petra; Walters, Fiona; Linxen, Sebastian; Gröhbiel, Urs

2017-01-01

This study analyses the use of a group space on the social networking site Facebook as a way to facilitate research supervision for teams of learners. Borrowing Lee's framework for research supervision, the goal was to understand how supervision and learning was achieved in, and shaped by, the properties of a social networking space. For this…

Speaker emotion recognition: from classical classifiers to deep neural networks

NASA Astrophysics Data System (ADS)

Mezghani, Eya; Charfeddine, Maha; Nicolas, Henri; Ben Amar, Chokri

2018-04-01

Speaker emotion recognition is considered among the most challenging tasks in recent years. In fact, automatic systems for security, medicine or education can be improved when considering the speech affective state. In this paper, a twofold approach for speech emotion classification is proposed. At the first side, a relevant set of features is adopted, and then at the second one, numerous supervised training techniques, involving classic methods as well as deep learning, are experimented. Experimental results indicate that deep architecture can improve classification performance on two affective databases, the Berlin Dataset of Emotional Speech and the SAVEE Dataset Surrey Audio-Visual Expressed Emotion.

Effect of filtration of signals of brain activity on quality of recognition of brain activity patterns using artificial intelligence methods

NASA Astrophysics Data System (ADS)

Hramov, Alexander E.; Frolov, Nikita S.; Musatov, Vyachaslav Yu.

2018-02-01

In present work we studied features of the human brain states classification, corresponding to the real movements of hands and legs. For this purpose we used supervised learning algorithm based on feed-forward artificial neural networks (ANNs) with error back-propagation along with the support vector machine (SVM) method. We compared the quality of operator movements classification by means of EEG signals obtained experimentally in the absence of preliminary processing and after filtration in different ranges up to 25 Hz. It was shown that low-frequency filtering of multichannel EEG data significantly improved accuracy of operator movements classification.

An equation-of-state-meter of quantum chromodynamics transition from deep learning.

PubMed

Pang, Long-Gang; Zhou, Kai; Su, Nan; Petersen, Hannah; Stöcker, Horst; Wang, Xin-Nian

2018-01-15

A primordial state of matter consisting of free quarks and gluons that existed in the early universe a few microseconds after the Big Bang is also expected to form in high-energy heavy-ion collisions. Determining the equation of state (EoS) of such a primordial matter is the ultimate goal of high-energy heavy-ion experiments. Here we use supervised learning with a deep convolutional neural network to identify the EoS employed in the relativistic hydrodynamic simulations of heavy ion collisions. High-level correlations of particle spectra in transverse momentum and azimuthal angle learned by the network act as an effective EoS-meter in deciphering the nature of the phase transition in quantum chromodynamics. Such EoS-meter is model-independent and insensitive to other simulation inputs including the initial conditions for hydrodynamic simulations.

A Deep Machine Learning Method for Classifying Cyclic Time Series of Biological Signals Using Time-Growing Neural Network.

PubMed

Gharehbaghi, Arash; Linden, Maria

2017-10-12

This paper presents a novel method for learning the cyclic contents of stochastic time series: the deep time-growing neural network (DTGNN). The DTGNN combines supervised and unsupervised methods in different levels of learning for an enhanced performance. It is employed by a multiscale learning structure to classify cyclic time series (CTS), in which the dynamic contents of the time series are preserved in an efficient manner. This paper suggests a systematic procedure for finding the design parameter of the classification method for a one-versus-multiple class application. A novel validation method is also suggested for evaluating the structural risk, both in a quantitative and a qualitative manner. The effect of the DTGNN on the performance of the classifier is statistically validated through the repeated random subsampling using different sets of CTS, from different medical applications. The validation involves four medical databases, comprised of 108 recordings of the electroencephalogram signal, 90 recordings of the electromyogram signal, 130 recordings of the heart sound signal, and 50 recordings of the respiratory sound signal. Results of the statistical validations show that the DTGNN significantly improves the performance of the classification and also exhibits an optimal structural risk.

Application of Deep Learning to Detect Precursors of Tropical Cyclone

NASA Astrophysics Data System (ADS)

Matsuoka, D.; Nakano, M.; Sugiyama, D.; Uchida, S.

2017-12-01

Tropical cyclones (TCs) affect significant damage to human society. Predicting TC generation as soon as possible is important issue in both academic and social perspectives. In the present work, we investigate the probability of predicting TCs seven days prior using deep neural networks. The training data is produced from 30-year cloud resolving global atmospheric simulation (NICAM) with 14 km horizontal resolution (Kodama et al., 2015). We employed a TCs tracking algorithm (Sugi et al., 2002; Nakano et al., 2015) to NICAM simulation data in order to generate supervised cloud images (horizontal sizes are 800-1,000km). We generate approximately one million images of "TCs (include their precursors)" and "not TCs (low pressure clouds)". We generate ten types of image classifier based on 2-dimensional convolutional neural network, includes four convolutional layers, three pooling layers and two fully connected layers. The final predicted results are obtained by these ensemble mean values. Generated classifiers are applied to untrained global simulation data (four million test images). As a result, we succeeded in predicting the precursors of TCs seven and five days before their formation with a Recall of 88.6% and 89.6% (Precision is 11.4%), respectively.

Quantification of the progression of CMV infection as observed from retinal angiograms in patients with AIDS

NASA Astrophysics Data System (ADS)

Brahmi, Djamel; Cassoux, Nathalie; Serruys, Camille; Giron, Alain; Lehoang, Phuc; Fertil, Bernard

1999-05-01

To support ophthalmologists in their daily routine and enable the quantitative assessment of progression of Cytomegalovirus infection as observed on series of retinal angiograms, a methodology allowing an accurate comparison of retinal borders has been developed. In order to evaluate accuracy of borders, ophthalmologists have been asked to repeatedly outline boundaries between infected and noninfected areas. As a matter of fact, accuracy of drawing relies on local features such as contrast, quality of image, background..., all factors which make the boundaries more or less perceptible from one part of an image to another. In order to directly estimate accuracy of retinal border from image analysis, an artificial neural network (a succession of unsupervised and supervised neural networks) has been designed to correlate accuracy of drawing (as calculated form ophthalmologists' hand-outlines) with local features of the underlying image. Our method has been applied to the quantification of CMV retinitis. It is shown that accuracy of border is properly predicted and characterized by a confident envelope that allows, after a registration phase based on fixed landmarks such as vessel forks, to accurately assess the evolution of CMV infection.

Taming Many-Parameter BSM Models with Bayesian Neural Networks

NASA Astrophysics Data System (ADS)

Kuchera, M. P.; Karbo, A.; Prosper, H. B.; Sanchez, A.; Taylor, J. Z.

2017-09-01

The search for physics Beyond the Standard Model (BSM) is a major focus of large-scale high energy physics experiments. One method is to look for specific deviations from the Standard Model that are predicted by BSM models. In cases where the model has a large number of free parameters, standard search methods become intractable due to computation time. This talk presents results using Bayesian Neural Networks, a supervised machine learning method, to enable the study of higher-dimensional models. The popular phenomenological Minimal Supersymmetric Standard Model was studied as an example of the feasibility and usefulness of this method. Graphics Processing Units (GPUs) are used to expedite the calculations. Cross-section predictions for 13 TeV proton collisions will be presented. My participation in the Conference Experience for Undergraduates (CEU) in 2004-2006 exposed me to the national and global significance of cutting-edge research. At the 2005 CEU, I presented work from the previous summer's SULI internship at Lawrence Berkeley Laboratory, where I learned to program while working on the Majorana Project. That work inspired me to follow a similar research path, which led me to my current work on computational methods applied to BSM physics.

A novel and reliable computational intelligence system for breast cancer detection.

PubMed

Zadeh Shirazi, Amin; Seyyed Mahdavi Chabok, Seyyed Javad; Mohammadi, Zahra

2018-05-01

Cancer is the second important morbidity and mortality factor among women and the most incident type is breast cancer. This paper suggests a hybrid computational intelligence model based on unsupervised and supervised learning techniques, i.e., self-organizing map (SOM) and complex-valued neural network (CVNN), for reliable detection of breast cancer. The dataset used in this paper consists of 822 patients with five features (patient's breast mass shape, margin, density, patient's age, and Breast Imaging Reporting and Data System assessment). The proposed model was used for the first time and can be categorized in two stages. In the first stage, considering the input features, SOM technique was used to cluster the patients with the most similarity. Then, in the second stage, for each cluster, the patient's features were applied to complex-valued neural network and dealt with to classify breast cancer severity (benign or malign). The obtained results corresponding to each patient were compared to the medical diagnosis results using receiver operating characteristic analyses and confusion matrix. In the testing phase, health and disease detection ratios were 94 and 95%, respectively. Accordingly, the superiority of the proposed model was proved and can be used for reliable and robust detection of breast cancer.

Neural networks for satellite remote sensing and robotic sensor interpretation

NASA Astrophysics Data System (ADS)

Martens, Siegfried

Remote sensing of forests and robotic sensor fusion can be viewed, in part, as supervised learning problems, mapping from sensory input to perceptual output. This dissertation develops ARTMAP neural networks for real-time category learning, pattern recognition, and prediction tailored to remote sensing and robotics applications. Three studies are presented. The first two use ARTMAP to create maps from remotely sensed data, while the third uses an ARTMAP system for sensor fusion on a mobile robot. The first study uses ARTMAP to predict vegetation mixtures in the Plumas National Forest based on spectral data from the Landsat Thematic Mapper satellite. While most previous ARTMAP systems have predicted discrete output classes, this project develops new capabilities for multi-valued prediction. On the mixture prediction task, the new network is shown to perform better than maximum likelihood and linear mixture models. The second remote sensing study uses an ARTMAP classification system to evaluate the relative importance of spectral and terrain data for map-making. This project has produced a large-scale map of remotely sensed vegetation in the Sierra National Forest. Network predictions are validated with ground truth data, and maps produced using the ARTMAP system are compared to a map produced by human experts. The ARTMAP Sierra map was generated in an afternoon, while the labor intensive expert method required nearly a year to perform the same task. The robotics research uses an ARTMAP system to integrate visual information and ultrasonic sensory information on a B14 mobile robot. The goal is to produce a more accurate measure of distance than is provided by the raw sensors. ARTMAP effectively combines sensory sources both within and between modalities. The improved distance percept is used to produce occupancy grid visualizations of the robot's environment. The maps produced point to specific problems of raw sensory information processing and demonstrate the benefits of using a neural network system for sensor fusion.

Classifying social anxiety disorder using multivoxel pattern analyses of brain function and structure☆

PubMed Central

Frick, Andreas; Gingnell, Malin; Marquand, Andre F.; Howner, Katarina; Fischer, Håkan; Kristiansson, Marianne; Williams, Steven C.R.; Fredrikson, Mats; Furmark, Tomas

2014-01-01

Functional neuroimaging of social anxiety disorder (SAD) support altered neural activation to threat-provoking stimuli focally in the fear network, while structural differences are distributed over the temporal and frontal cortices as well as limbic structures. Previous neuroimaging studies have investigated the brain at the voxel level using mass-univariate methods which do not enable detection of more complex patterns of activity and structural alterations that may separate SAD from healthy individuals. Support vector machine (SVM) is a supervised machine learning method that capitalizes on brain activation and structural patterns to classify individuals. The aim of this study was to investigate if it is possible to discriminate SAD patients (n = 14) from healthy controls (n = 12) using SVM based on (1) functional magnetic resonance imaging during fearful face processing and (2) regional gray matter volume. Whole brain and region of interest (fear network) SVM analyses were performed for both modalities. For functional scans, significant classifications were obtained both at whole brain level and when restricting the analysis to the fear network while gray matter SVM analyses correctly classified participants only when using the whole brain search volume. These results support that SAD is characterized by aberrant neural activation to affective stimuli in the fear network, while disorder-related alterations in regional gray matter volume are more diffusely distributed over the whole brain. SVM may thus be useful for identifying imaging biomarkers of SAD. PMID:24239689

A Supervised Approach to Windowing Detection on Dynamic Networks

DTIC Science & Technology

2017-07-01

A supervised approach to windowing detection on dynamic networks Benjamin Fish University of Illinois at Chicago 1200 W. Harrison St. Chicago...Using this framework, we introduce windowing algorithms that take a supervised approach : they leverage ground truth on training data to find a good...windowing of the test data. We compare the supervised approach to previous approaches and several baselines on real data. ACM Reference format: Benjamin

A model of traffic signs recognition with convolutional neural network

NASA Astrophysics Data System (ADS)

Hu, Haihe; Li, Yujian; Zhang, Ting; Huo, Yi; Kuang, Wenqing

2016-10-01

In real traffic scenes, the quality of captured images are generally low due to some factors such as lighting conditions, and occlusion on. All of these factors are challengeable for automated recognition algorithms of traffic signs. Deep learning has provided a new way to solve this kind of problems recently. The deep network can automatically learn features from a large number of data samples and obtain an excellent recognition performance. We therefore approach this task of recognition of traffic signs as a general vision problem, with few assumptions related to road signs. We propose a model of Convolutional Neural Network (CNN) and apply the model to the task of traffic signs recognition. The proposed model adopts deep CNN as the supervised learning model, directly takes the collected traffic signs image as the input, alternates the convolutional layer and subsampling layer, and automatically extracts the features for the recognition of the traffic signs images. The proposed model includes an input layer, three convolutional layers, three subsampling layers, a fully-connected layer, and an output layer. To validate the proposed model, the experiments are implemented using the public dataset of China competition of fuzzy image processing. Experimental results show that the proposed model produces a recognition accuracy of 99.01 % on the training dataset, and yield a record of 92% on the preliminary contest within the fourth best.

Competitive Deep-Belief Networks for Underwater Acoustic Target Recognition

PubMed Central

Shen, Sheng; Yao, Xiaohui; Sheng, Meiping; Wang, Chen

2018-01-01

Underwater acoustic target recognition based on ship-radiated noise belongs to the small-sample-size recognition problems. A competitive deep-belief network is proposed to learn features with more discriminative information from labeled and unlabeled samples. The proposed model consists of four stages: (1) A standard restricted Boltzmann machine is pretrained using a large number of unlabeled data to initialize its parameters; (2) the hidden units are grouped according to categories, which provides an initial clustering model for competitive learning; (3) competitive training and back-propagation algorithms are used to update the parameters to accomplish the task of clustering; (4) by applying layer-wise training and supervised fine-tuning, a deep neural network is built to obtain features. Experimental results show that the proposed method can achieve classification accuracy of 90.89%, which is 8.95% higher than the accuracy obtained by the compared methods. In addition, the highest accuracy of our method is obtained with fewer features than other methods. PMID:29570642

Machine learning vortices at the Kosterlitz-Thouless transition

NASA Astrophysics Data System (ADS)

Beach, Matthew J. S.; Golubeva, Anna; Melko, Roger G.

2018-01-01

Efficient and automated classification of phases from minimally processed data is one goal of machine learning in condensed-matter and statistical physics. Supervised algorithms trained on raw samples of microstates can successfully detect conventional phase transitions via learning a bulk feature such as an order parameter. In this paper, we investigate whether neural networks can learn to classify phases based on topological defects. We address this question on the two-dimensional classical XY model which exhibits a Kosterlitz-Thouless transition. We find significant feature engineering of the raw spin states is required to convincingly claim that features of the vortex configurations are responsible for learning the transition temperature. We further show a single-layer network does not correctly classify the phases of the XY model, while a convolutional network easily performs classification by learning the global magnetization. Finally, we design a deep network capable of learning vortices without feature engineering. We demonstrate the detection of vortices does not necessarily result in the best classification accuracy, especially for lattices of less than approximately 1000 spins. For larger systems, it remains a difficult task to learn vortices.

“We make choices we think are going to save us”: Debate and stance identification for online breast cancer CAM discussions

PubMed Central

Zhang, Shaodian; Qiu, Lin; Chen, Frank; Zhang, Weinan; Yu, Yong; Elhadad, Noémie

2017-01-01

Patients discuss complementary and alternative medicine (CAM) in online health communities. Sometimes, patients’ conflicting opinions toward CAM-related issues trigger debates in the community. The objectives of this paper are to identify such debates, identify controversial CAM therapies in a popular online breast cancer community, as well as patients’ stances towards them. To scale our analysis, we trained a set of classifiers. We first constructed a supervised classifier based on a long short-term memory neural network (LSTM) stacked over a convolutional neural network (CNN) to detect automatically CAM-related debates from a popular breast cancer forum. Members’ stances in these debates were also identified by a CNN-based classifier. Finally, posts automatically flagged as debates by the classifier were analyzed to explore which specific CAM therapies trigger debates more often than others. Our methods are able to detect CAM debates with F score of 77%, and identify stances with F score of 70%. The debate classifier identified about 1/6 of all CAM-related posts as debate. About 60% of CAM-related debate posts represent the supportive stance toward CAM usage. Qualitative analysis shows that some specific therapies, such as Gerson therapy and usage of laetrile, trigger debates frequently among members of the breast cancer community. This study demonstrates that neural networks can effectively locate debates on usage and effectiveness of controversial CAM therapies, and can help make sense of patients’ opinions on such issues under dispute. As to CAM for breast cancer, perceptions of their effectiveness vary among patients. Many of the specific therapies trigger debates frequently and are worth more exploration in future work. PMID:28967000

"We make choices we think are going to save us": Debate and stance identification for online breast cancer CAM discussions.

PubMed

Zhang, Shaodian; Qiu, Lin; Chen, Frank; Zhang, Weinan; Yu, Yong; Elhadad, Noémie

2017-04-01

Patients discuss complementary and alternative medicine (CAM) in online health communities. Sometimes, patients' conflicting opinions toward CAM-related issues trigger debates in the community. The objectives of this paper are to identify such debates, identify controversial CAM therapies in a popular online breast cancer community, as well as patients' stances towards them. To scale our analysis, we trained a set of classifiers. We first constructed a supervised classifier based on a long short-term memory neural network (LSTM) stacked over a convolutional neural network (CNN) to detect automatically CAM-related debates from a popular breast cancer forum. Members' stances in these debates were also identified by a CNN-based classifier. Finally, posts automatically flagged as debates by the classifier were analyzed to explore which specific CAM therapies trigger debates more often than others. Our methods are able to detect CAM debates with F score of 77%, and identify stances with F score of 70%. The debate classifier identified about 1/6 of all CAM-related posts as debate. About 60% of CAM-related debate posts represent the supportive stance toward CAM usage. Qualitative analysis shows that some specific therapies, such as Gerson therapy and usage of laetrile, trigger debates frequently among members of the breast cancer community. This study demonstrates that neural networks can effectively locate debates on usage and effectiveness of controversial CAM therapies, and can help make sense of patients' opinions on such issues under dispute. As to CAM for breast cancer, perceptions of their effectiveness vary among patients. Many of the specific therapies trigger debates frequently and are worth more exploration in future work.

Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning.

PubMed

Shin, Hoo-Chang; Roth, Holger R; Gao, Mingchen; Lu, Le; Xu, Ziyue; Nogues, Isabella; Yao, Jianhua; Mollura, Daniel; Summers, Ronald M

2016-05-01

Remarkable progress has been made in image recognition, primarily due to the availability of large-scale annotated datasets and deep convolutional neural networks (CNNs). CNNs enable learning data-driven, highly representative, hierarchical image features from sufficient training data. However, obtaining datasets as comprehensively annotated as ImageNet in the medical imaging domain remains a challenge. There are currently three major techniques that successfully employ CNNs to medical image classification: training the CNN from scratch, using off-the-shelf pre-trained CNN features, and conducting unsupervised CNN pre-training with supervised fine-tuning. Another effective method is transfer learning, i.e., fine-tuning CNN models pre-trained from natural image dataset to medical image tasks. In this paper, we exploit three important, but previously understudied factors of employing deep convolutional neural networks to computer-aided detection problems. We first explore and evaluate different CNN architectures. The studied models contain 5 thousand to 160 million parameters, and vary in numbers of layers. We then evaluate the influence of dataset scale and spatial image context on performance. Finally, we examine when and why transfer learning from pre-trained ImageNet (via fine-tuning) can be useful. We study two specific computer-aided detection (CADe) problems, namely thoraco-abdominal lymph node (LN) detection and interstitial lung disease (ILD) classification. We achieve the state-of-the-art performance on the mediastinal LN detection, and report the first five-fold cross-validation classification results on predicting axial CT slices with ILD categories. Our extensive empirical evaluation, CNN model analysis and valuable insights can be extended to the design of high performance CAD systems for other medical imaging tasks.

Combining deep residual neural network features with supervised machine learning algorithms to classify diverse food image datasets.

PubMed

McAllister, Patrick; Zheng, Huiru; Bond, Raymond; Moorhead, Anne

2018-04-01

Obesity is increasing worldwide and can cause many chronic conditions such as type-2 diabetes, heart disease, sleep apnea, and some cancers. Monitoring dietary intake through food logging is a key method to maintain a healthy lifestyle to prevent and manage obesity. Computer vision methods have been applied to food logging to automate image classification for monitoring dietary intake. In this work we applied pretrained ResNet-152 and GoogleNet convolutional neural networks (CNNs), initially trained using ImageNet Large Scale Visual Recognition Challenge (ILSVRC) dataset with MatConvNet package, to extract features from food image datasets; Food 5K, Food-11, RawFooT-DB, and Food-101. Deep features were extracted from CNNs and used to train machine learning classifiers including artificial neural network (ANN), support vector machine (SVM), Random Forest, and Naive Bayes. Results show that using ResNet-152 deep features with SVM with RBF kernel can accurately detect food items with 99.4% accuracy using Food-5K validation food image dataset and 98.8% with Food-5K evaluation dataset using ANN, SVM-RBF, and Random Forest classifiers. Trained with ResNet-152 features, ANN can achieve 91.34%, 99.28% when applied to Food-11 and RawFooT-DB food image datasets respectively and SVM with RBF kernel can achieve 64.98% with Food-101 image dataset. From this research it is clear that using deep CNN features can be used efficiently for diverse food item image classification. The work presented in this research shows that pretrained ResNet-152 features provide sufficient generalisation power when applied to a range of food image classification tasks. Copyright © 2018 Elsevier Ltd. All rights reserved.

Ellipsoidal fuzzy learning for smart car platoons

NASA Astrophysics Data System (ADS)

Dickerson, Julie A.; Kosko, Bart

1993-12-01

A neural-fuzzy system combined supervised and unsupervised learning to find and tune the fuzzy-rules. An additive fuzzy system approximates a function by covering its graph with fuzzy rules. A fuzzy rule patch can take the form of an ellipsoid in the input-output space. Unsupervised competitive learning found the statistics of data clusters. The covariance matrix of each synaptic quantization vector defined on ellipsoid centered at the centroid of the data cluster. Tightly clustered data gave smaller ellipsoids or more certain rules. Sparse data gave larger ellipsoids or less certain rules. Supervised learning tuned the ellipsoids to improve the approximation. The supervised neural system used gradient descent to find the ellipsoidal fuzzy patches. It locally minimized the mean-squared error of the fuzzy approximation. Hybrid ellipsoidal learning estimated the control surface for a smart car controller.

An equation-of-state-meter of quantum chromodynamics transition from deep learning

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

Pang, Long-Gang; Zhou, Kai; Su, Nan

A primordial state of matter consisting of free quarks and gluons that existed in the early universe a few microseconds after the Big Bang is also expected to form in high-energy heavy-ion collisions. Determining the equation of state (EoS) of such a primordial matter is the ultimate goal of high-energy heavy-ion experiments. Here we use supervised learning with a deep convolutional neural network to identify the EoS employed in the relativistic hydrodynamic simulations of heavy ion collisions. High-level correlations of particle spectra in transverse momentum and azimuthal angle learned by the network act as an effective EoS-meter in deciphering themore » nature of the phase transition in quantum chromodynamics. Finally, such EoS-meter is model-independent and insensitive to other simulation inputs including the initial conditions for hydrodynamic simulations.« less

An equation-of-state-meter of quantum chromodynamics transition from deep learning

DOE PAGES

Pang, Long-Gang; Zhou, Kai; Su, Nan; ...

2018-01-15

A primordial state of matter consisting of free quarks and gluons that existed in the early universe a few microseconds after the Big Bang is also expected to form in high-energy heavy-ion collisions. Determining the equation of state (EoS) of such a primordial matter is the ultimate goal of high-energy heavy-ion experiments. Here we use supervised learning with a deep convolutional neural network to identify the EoS employed in the relativistic hydrodynamic simulations of heavy ion collisions. High-level correlations of particle spectra in transverse momentum and azimuthal angle learned by the network act as an effective EoS-meter in deciphering themore » nature of the phase transition in quantum chromodynamics. Finally, such EoS-meter is model-independent and insensitive to other simulation inputs including the initial conditions for hydrodynamic simulations.« less

Novel maximum-margin training algorithms for supervised neural networks.

PubMed

Ludwig, Oswaldo; Nunes, Urbano

2010-06-01

This paper proposes three novel training methods, two of them based on the backpropagation approach and a third one based on information theory for multilayer perceptron (MLP) binary classifiers. Both backpropagation methods are based on the maximal-margin (MM) principle. The first one, based on the gradient descent with adaptive learning rate algorithm (GDX) and named maximum-margin GDX (MMGDX), directly increases the margin of the MLP output-layer hyperplane. The proposed method jointly optimizes both MLP layers in a single process, backpropagating the gradient of an MM-based objective function, through the output and hidden layers, in order to create a hidden-layer space that enables a higher margin for the output-layer hyperplane, avoiding the testing of many arbitrary kernels, as occurs in case of support vector machine (SVM) training. The proposed MM-based objective function aims to stretch out the margin to its limit. An objective function based on Lp-norm is also proposed in order to take into account the idea of support vectors, however, overcoming the complexity involved in solving a constrained optimization problem, usually in SVM training. In fact, all the training methods proposed in this paper have time and space complexities O(N) while usual SVM training methods have time complexity O(N (3)) and space complexity O(N (2)) , where N is the training-data-set size. The second approach, named minimization of interclass interference (MICI), has an objective function inspired on the Fisher discriminant analysis. Such algorithm aims to create an MLP hidden output where the patterns have a desirable statistical distribution. In both training methods, the maximum area under ROC curve (AUC) is applied as stop criterion. The third approach offers a robust training framework able to take the best of each proposed training method. The main idea is to compose a neural model by using neurons extracted from three other neural networks, each one previously trained by MICI, MMGDX, and Levenberg-Marquard (LM), respectively. The resulting neural network was named assembled neural network (ASNN). Benchmark data sets of real-world problems have been used in experiments that enable a comparison with other state-of-the-art classifiers. The results provide evidence of the effectiveness of our methods regarding accuracy, AUC, and balanced error rate.

Deep Logic Networks: Inserting and Extracting Knowledge From Deep Belief Networks.

PubMed

Tran, Son N; d'Avila Garcez, Artur S

2018-02-01

Developments in deep learning have seen the use of layerwise unsupervised learning combined with supervised learning for fine-tuning. With this layerwise approach, a deep network can be seen as a more modular system that lends itself well to learning representations. In this paper, we investigate whether such modularity can be useful to the insertion of background knowledge into deep networks, whether it can improve learning performance when it is available, and to the extraction of knowledge from trained deep networks, and whether it can offer a better understanding of the representations learned by such networks. To this end, we use a simple symbolic language-a set of logical rules that we call confidence rules-and show that it is suitable for the representation of quantitative reasoning in deep networks. We show by knowledge extraction that confidence rules can offer a low-cost representation for layerwise networks (or restricted Boltzmann machines). We also show that layerwise extraction can produce an improvement in the accuracy of deep belief networks. Furthermore, the proposed symbolic characterization of deep networks provides a novel method for the insertion of prior knowledge and training of deep networks. With the use of this method, a deep neural-symbolic system is proposed and evaluated, with the experimental results indicating that modularity through the use of confidence rules and knowledge insertion can be beneficial to network performance.

Benchmarking protein classification algorithms via supervised cross-validation.

PubMed

Kertész-Farkas, Attila; Dhir, Somdutta; Sonego, Paolo; Pacurar, Mircea; Netoteia, Sergiu; Nijveen, Harm; Kuzniar, Arnold; Leunissen, Jack A M; Kocsor, András; Pongor, Sándor

2008-04-24

Development and testing of protein classification algorithms are hampered by the fact that the protein universe is characterized by groups vastly different in the number of members, in average protein size, similarity within group, etc. Datasets based on traditional cross-validation (k-fold, leave-one-out, etc.) may not give reliable estimates on how an algorithm will generalize to novel, distantly related subtypes of the known protein classes. Supervised cross-validation, i.e., selection of test and train sets according to the known subtypes within a database has been successfully used earlier in conjunction with the SCOP database. Our goal was to extend this principle to other databases and to design standardized benchmark datasets for protein classification. Hierarchical classification trees of protein categories provide a simple and general framework for designing supervised cross-validation strategies for protein classification. Benchmark datasets can be designed at various levels of the concept hierarchy using a simple graph-theoretic distance. A combination of supervised and random sampling was selected to construct reduced size model datasets, suitable for algorithm comparison. Over 3000 new classification tasks were added to our recently established protein classification benchmark collection that currently includes protein sequence (including protein domains and entire proteins), protein structure and reading frame DNA sequence data. We carried out an extensive evaluation based on various machine-learning algorithms such as nearest neighbor, support vector machines, artificial neural networks, random forests and logistic regression, used in conjunction with comparison algorithms, BLAST, Smith-Waterman, Needleman-Wunsch, as well as 3D comparison methods DALI and PRIDE. The resulting datasets provide lower, and in our opinion more realistic estimates of the classifier performance than do random cross-validation schemes. A combination of supervised and random sampling was used to construct model datasets, suitable for algorithm comparison.

An Investigation of the Application of Artificial Neural Networks to Adaptive Optics Imaging Systems

DTIC Science & Technology

1991-12-01

neural network and the feedforward neural network studied is the single layer perceptron artificial neural network . The recurrent artificial neural network input...features are the wavefront sensor slope outputs and neighboring actuator feedback commands. The feedforward artificial neural network input

Learned filters for object detection in multi-object visual tracking

NASA Astrophysics Data System (ADS)

Stamatescu, Victor; Wong, Sebastien; McDonnell, Mark D.; Kearney, David

2016-05-01

We investigate the application of learned convolutional filters in multi-object visual tracking. The filters were learned in both a supervised and unsupervised manner from image data using artificial neural networks. This work follows recent results in the field of machine learning that demonstrate the use learned filters for enhanced object detection and classification. Here we employ a track-before-detect approach to multi-object tracking, where tracking guides the detection process. The object detection provides a probabilistic input image calculated by selecting from features obtained using banks of generative or discriminative learned filters. We present a systematic evaluation of these convolutional filters using a real-world data set that examines their performance as generic object detectors.

Classification of 2-dimensional array patterns: assembling many small neural networks is better than using a large one.

PubMed

Chen, Liang; Xue, Wei; Tokuda, Naoyuki

2010-08-01

In many pattern classification/recognition applications of artificial neural networks, an object to be classified is represented by a fixed sized 2-dimensional array of uniform type, which corresponds to the cells of a 2-dimensional grid of the same size. A general neural network structure, called an undistricted neural network, which takes all the elements in the array as inputs could be used for problems such as these. However, a districted neural network can be used to reduce the training complexity. A districted neural network usually consists of two levels of sub-neural networks. Each of the lower level neural networks, called a regional sub-neural network, takes the elements in a region of the array as its inputs and is expected to output a temporary class label, called an individual opinion, based on the partial information of the entire array. The higher level neural network, called an assembling sub-neural network, uses the outputs (opinions) of regional sub-neural networks as inputs, and by consensus derives the label decision for the object. Each of the sub-neural networks can be trained separately and thus the training is less expensive. The regional sub-neural networks can be trained and performed in parallel and independently, therefore a high speed can be achieved. We prove theoretically in this paper, using a simple model, that a districted neural network is actually more stable than an undistricted neural network in noisy environments. We conjecture that the result is valid for all neural networks. This theory is verified by experiments involving gender classification and human face recognition. We conclude that a districted neural network is highly recommended for neural network applications in recognition or classification of 2-dimensional array patterns in highly noisy environments. Copyright (c) 2010 Elsevier Ltd. All rights reserved.

Goal Directed Model Inversion: Learning Within Domain Constraints

NASA Technical Reports Server (NTRS)

Colombano, Silvano P.; Compton, Michael; Raghavan, Bharathi; Friedland, Peter (Technical Monitor)

1994-01-01

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired target. The fundamental driving mechanism of GDMI is to learn from success. Given that a wrong outcome is achieved, one notes that the action that produced that outcome "would have been right if the outcome had been the desired one." The algorithm makes use of these intermediate "successes" to achieve the final goal. A unique and potentially very important feature of this algorithm is the ability to modify the output of the learning module to force upon it a desired syntactic structure. This differs from ordinary supervised learning in the following way: in supervised learning the exact desired output pattern must be provided. In GDMI instead, it is possible to require simply that the output obey certain rules, i.e., that it "make sense" in some way determined by the knowledge domain. The exact pattern that will achieve the desired outcome is then found by the system. The ability to impose rules while allowing the system to search for its own answers in the context of neural networks is potentially a major breakthrough in two ways: (1) it may allow the construction of networks that can incorporate immediately some important knowledge, i.e., would not need to learn everything from scratch as normally required at present; and (2) learning and searching would be limited to the areas where it is necessary, thus facilitating and speeding up the process. These points are illustrated with examples from robotic path planning and parametric design.

Goal Directed Model Inversion: Learning Within Domain Constraints

NASA Technical Reports Server (NTRS)

Colombano, Silvano P.; Compton, Michael; Raghavan, Bharathi; Lum, Henry, Jr. (Technical Monitor)

1994-01-01

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired target. The fundamental driving mechanism of GDMI is to learn from success. Given that a wrong outcome is achieved, one notes that the action that produced that outcome "would have been right if the outcome had been the desired one." The algorithm makes use of these intermediate "successes" to achieve the final goal. A unique and potentially very important feature of this algorithm is the ability to modify the output of the learning module to force upon it a desired syntactic structure. This differs from ordinary supervised learning in the following way: in supervised learning the exact desired output pattern must be provided. In GDMI instead, it is possible to require simply that the output obey certain rules, i.e., that it "make sense" in some way determined by the knowledge domain. The exact pattern that will achieve the desired outcome is then found by the system. The ability to impose rules while allowing the system to search for its own answers in the context of neural networks is potentially a major breakthrough in two ways: 1) it may allow the construction of networks that can incorporate immediately some important knowledge, i.e. would not need to learn everything from scratch as normally required at present, and 2) learning and searching would be limited to the areas where it is necessary, thus facilitating and speeding up the process. These points are illustrated with examples from robotic path planning and parametric design.

Advanced Aeroservoelastic Testing and Data Analysis (Les Essais Aeroservoelastiques et l’Analyse des Donnees).

DTIC Science & Technology

1995-11-01

network - based AFS concepts. Neural networks can addition of vanes in each engine exhaust for thrust provide...parameter estimation programs 19-11 8.6 Neural Network Based Methods unknown parameters of the postulated state space model Artificial neural network ...Forward Neural Network the network that the applicability of the recurrent neural and ii) Recurrent Neural Network [117-119]. network to

Towards autonomous neuroprosthetic control using Hebbian reinforcement learning.

PubMed

Mahmoudi, Babak; Pohlmeyer, Eric A; Prins, Noeline W; Geng, Shijia; Sanchez, Justin C

2013-12-01

Our goal was to design an adaptive neuroprosthetic controller that could learn the mapping from neural states to prosthetic actions and automatically adjust adaptation using only a binary evaluative feedback as a measure of desirability/undesirability of performance. Hebbian reinforcement learning (HRL) in a connectionist network was used for the design of the adaptive controller. The method combines the efficiency of supervised learning with the generality of reinforcement learning. The convergence properties of this approach were studied using both closed-loop control simulations and open-loop simulations that used primate neural data from robot-assisted reaching tasks. The HRL controller was able to perform classification and regression tasks using its episodic and sequential learning modes, respectively. In our experiments, the HRL controller quickly achieved convergence to an effective control policy, followed by robust performance. The controller also automatically stopped adapting the parameters after converging to a satisfactory control policy. Additionally, when the input neural vector was reorganized, the controller resumed adaptation to maintain performance. By estimating an evaluative feedback directly from the user, the HRL control algorithm may provide an efficient method for autonomous adaptation of neuroprosthetic systems. This method may enable the user to teach the controller the desired behavior using only a simple feedback signal.

Neural networks for aircraft control

NASA Technical Reports Server (NTRS)

Linse, Dennis

1990-01-01

Current research in Artificial Neural Networks indicates that networks offer some potential advantages in adaptation and fault tolerance. This research is directed at determining the possible applicability of neural networks to aircraft control. The first application will be to aircraft trim. Neural network node characteristics, network topology and operation, neural network learning and example histories using neighboring optimal control with a neural net are discussed.

Clinical Relation Extraction Toward Drug Safety Surveillance Using Electronic Health Record Narratives: Classical Learning Versus Deep Learning.

PubMed

Munkhdalai, Tsendsuren; Liu, Feifan; Yu, Hong

2018-04-25

Medication and adverse drug event (ADE) information extracted from electronic health record (EHR) notes can be a rich resource for drug safety surveillance. Existing observational studies have mainly relied on structured EHR data to obtain ADE information; however, ADEs are often buried in the EHR narratives and not recorded in structured data. To unlock ADE-related information from EHR narratives, there is a need to extract relevant entities and identify relations among them. In this study, we focus on relation identification. This study aimed to evaluate natural language processing and machine learning approaches using the expert-annotated medical entities and relations in the context of drug safety surveillance, and investigate how different learning approaches perform under different configurations. We have manually annotated 791 EHR notes with 9 named entities (eg, medication, indication, severity, and ADEs) and 7 different types of relations (eg, medication-dosage, medication-ADE, and severity-ADE). Then, we explored 3 supervised machine learning systems for relation identification: (1) a support vector machines (SVM) system, (2) an end-to-end deep neural network system, and (3) a supervised descriptive rule induction baseline system. For the neural network system, we exploited the state-of-the-art recurrent neural network (RNN) and attention models. We report the performance by macro-averaged precision, recall, and F1-score across the relation types. Our results show that the SVM model achieved the best average F1-score of 89.1% on test data, outperforming the long short-term memory (LSTM) model with attention (F1-score of 65.72%) as well as the rule induction baseline system (F1-score of 7.47%) by a large margin. The bidirectional LSTM model with attention achieved the best performance among different RNN models. With the inclusion of additional features in the LSTM model, its performance can be boosted to an average F1-score of 77.35%. It shows that classical learning models (SVM) remains advantageous over deep learning models (RNN variants) for clinical relation identification, especially for long-distance intersentential relations. However, RNNs demonstrate a great potential of significant improvement if more training data become available. Our work is an important step toward mining EHRs to improve the efficacy of drug safety surveillance. Most importantly, the annotated data used in this study will be made publicly available, which will further promote drug safety research in the community. ©Tsendsuren Munkhdalai, Feifan Liu, Hong Yu. Originally published in JMIR Public Health and Surveillance (http://publichealth.jmir.org), 25.04.2018.

Clinical Relation Extraction Toward Drug Safety Surveillance Using Electronic Health Record Narratives: Classical Learning Versus Deep Learning

PubMed Central

Munkhdalai, Tsendsuren; Liu, Feifan

2018-01-01

Background Medication and adverse drug event (ADE) information extracted from electronic health record (EHR) notes can be a rich resource for drug safety surveillance. Existing observational studies have mainly relied on structured EHR data to obtain ADE information; however, ADEs are often buried in the EHR narratives and not recorded in structured data. Objective To unlock ADE-related information from EHR narratives, there is a need to extract relevant entities and identify relations among them. In this study, we focus on relation identification. This study aimed to evaluate natural language processing and machine learning approaches using the expert-annotated medical entities and relations in the context of drug safety surveillance, and investigate how different learning approaches perform under different configurations. Methods We have manually annotated 791 EHR notes with 9 named entities (eg, medication, indication, severity, and ADEs) and 7 different types of relations (eg, medication-dosage, medication-ADE, and severity-ADE). Then, we explored 3 supervised machine learning systems for relation identification: (1) a support vector machines (SVM) system, (2) an end-to-end deep neural network system, and (3) a supervised descriptive rule induction baseline system. For the neural network system, we exploited the state-of-the-art recurrent neural network (RNN) and attention models. We report the performance by macro-averaged precision, recall, and F1-score across the relation types. Results Our results show that the SVM model achieved the best average F1-score of 89.1% on test data, outperforming the long short-term memory (LSTM) model with attention (F1-score of 65.72%) as well as the rule induction baseline system (F1-score of 7.47%) by a large margin. The bidirectional LSTM model with attention achieved the best performance among different RNN models. With the inclusion of additional features in the LSTM model, its performance can be boosted to an average F1-score of 77.35%. Conclusions It shows that classical learning models (SVM) remains advantageous over deep learning models (RNN variants) for clinical relation identification, especially for long-distance intersentential relations. However, RNNs demonstrate a great potential of significant improvement if more training data become available. Our work is an important step toward mining EHRs to improve the efficacy of drug safety surveillance. Most importantly, the annotated data used in this study will be made publicly available, which will further promote drug safety research in the community. PMID:29695376

Analysis of the changes in the tarcrete layer on the desert surface of Kuwait using satellite imagery and cell-based modeling

NASA Astrophysics Data System (ADS)

Al-Doasari, Ahmad E.

The 1991 Gulf War caused massive environmental damage in Kuwait. Deposition of oil and soot droplets from hundreds of burning oil-wells created a layer of tarcrete on the desert surface covering over 900 km2. This research investigates the spatial change in the tarcrete extent from 1991 to 1998 using Landsat Thematic Mapper (TM) imagery and statistical modeling techniques. The pixel structure of TM data allows the spatial analysis of the change in tarcrete extent to be conducted at the pixel (cell) level within a geographical information system (GIS). There are two components to this research. The first is a comparison of three remote sensing classification techniques used to map the tarcrete layer. The second is a spatial-temporal analysis and simulation of tarcrete changes through time. The analysis focuses on an area of 389 km2 located south of the Al-Burgan oil field. Five TM images acquired in 1991, 1993, 1994, 1995, and 1998 were geometrically and atmospherically corrected. These images were classified into six classes: oil lakes; heavy, intermediate, light, and traces of tarcrete; and sand. The classification methods tested were unsupervised, supervised, and neural network supervised (fuzzy ARTMAP). Field data of tarcrete characteristics were collected to support the classification process and to evaluate the classification accuracies. Overall, the neural network method is more accurate (60 percent) than the other two methods; both the unsupervised and the supervised classification accuracy assessments resulted in 46 percent accuracy. The five classifications were used in a lagged autologistic model to analyze the spatial changes of the tarcrete through time. The autologistic model correctly identified overall tarcrete contraction between 1991--1993 and 1995--1998. However, tarcrete contraction between 1993--1994 and 1994--1995 was less well marked, in part because of classification errors in the maps from these time periods. Initial simulations of tarcrete contraction with a cellular automaton model were not very successful. However, more accurate classifications could improve the simulations. This study illustrates how an empirical investigation using satellite images, field data, GIS, and spatial statistics can simulate dynamic land-cover change through the use of a discrete statistical and cellular automaton model.

Time Series Neural Network Model for Part-of-Speech Tagging Indonesian Language

NASA Astrophysics Data System (ADS)

Tanadi, Theo

2018-03-01

Part-of-speech tagging (POS tagging) is an important part in natural language processing. Many methods have been used to do this task, including neural network. This paper models a neural network that attempts to do POS tagging. A time series neural network is modelled to solve the problems that a basic neural network faces when attempting to do POS tagging. In order to enable the neural network to have text data input, the text data will get clustered first using Brown Clustering, resulting a binary dictionary that the neural network can use. To further the accuracy of the neural network, other features such as the POS tag, suffix, and affix of previous words would also be fed to the neural network.

Comparative Study on Interaction of Form and Motion Processing Streams by Applying Two Different Classifiers in Mechanism for Recognition of Biological Movement

PubMed Central

2014-01-01

Research on psychophysics, neurophysiology, and functional imaging shows particular representation of biological movements which contains two pathways. The visual perception of biological movements formed through the visual system called dorsal and ventral processing streams. Ventral processing stream is associated with the form information extraction; on the other hand, dorsal processing stream provides motion information. Active basic model (ABM) as hierarchical representation of the human object had revealed novelty in form pathway due to applying Gabor based supervised object recognition method. It creates more biological plausibility along with similarity with original model. Fuzzy inference system is used for motion pattern information in motion pathway creating more robustness in recognition process. Besides, interaction of these paths is intriguing and many studies in various fields considered it. Here, the interaction of the pathways to get more appropriated results has been investigated. Extreme learning machine (ELM) has been implied for classification unit of this model, due to having the main properties of artificial neural networks, but crosses from the difficulty of training time substantially diminished in it. Here, there will be a comparison between two different configurations, interactions using synergetic neural network and ELM, in terms of accuracy and compatibility. PMID:25276860

Evaluating the statistical performance of less applied algorithms in classification of worldview-3 imagery data in an urbanized landscape

NASA Astrophysics Data System (ADS)

Ranaie, Mehrdad; Soffianian, Alireza; Pourmanafi, Saeid; Mirghaffari, Noorollah; Tarkesh, Mostafa

2018-03-01

In recent decade, analyzing the remotely sensed imagery is considered as one of the most common and widely used procedures in the environmental studies. In this case, supervised image classification techniques play a central role. Hence, taking a high resolution Worldview-3 over a mixed urbanized landscape in Iran, three less applied image classification methods including Bagged CART, Stochastic gradient boosting model and Neural network with feature extraction were tested and compared with two prevalent methods: random forest and support vector machine with linear kernel. To do so, each method was run ten time and three validation techniques was used to estimate the accuracy statistics consist of cross validation, independent validation and validation with total of train data. Moreover, using ANOVA and Tukey test, statistical difference significance between the classification methods was significantly surveyed. In general, the results showed that random forest with marginal difference compared to Bagged CART and stochastic gradient boosting model is the best performing method whilst based on independent validation there was no significant difference between the performances of classification methods. It should be finally noted that neural network with feature extraction and linear support vector machine had better processing speed than other.

Assessing the performance of multiple spectral-spatial features of a hyperspectral image for classification of urban land cover classes using support vector machines and artificial neural network

NASA Astrophysics Data System (ADS)

Pullanagari, Reddy; Kereszturi, Gábor; Yule, Ian J.; Ghamisi, Pedram

2017-04-01

Accurate and spatially detailed mapping of complex urban environments is essential for land managers. Classifying high spectral and spatial resolution hyperspectral images is a challenging task because of its data abundance and computational complexity. Approaches with a combination of spectral and spatial information in a single classification framework have attracted special attention because of their potential to improve the classification accuracy. We extracted multiple features from spectral and spatial domains of hyperspectral images and evaluated them with two supervised classification algorithms; support vector machines (SVM) and an artificial neural network. The spatial features considered are produced by a gray level co-occurrence matrix and extended multiattribute profiles. All of these features were stacked, and the most informative features were selected using a genetic algorithm-based SVM. After selecting the most informative features, the classification model was integrated with a segmentation map derived using a hidden Markov random field. We tested the proposed method on a real application of a hyperspectral image acquired from AisaFENIX and on widely used hyperspectral images. From the results, it can be concluded that the proposed framework significantly improves the results with different spectral and spatial resolutions over different instrumentation.

Neural networks for the generation of sea bed models using airborne lidar bathymetry data

NASA Astrophysics Data System (ADS)

Kogut, Tomasz; Niemeyer, Joachim; Bujakiewicz, Aleksandra

2016-06-01

Various sectors of the economy such as transport and renewable energy have shown great interest in sea bed models. The required measurements are usually carried out by ship-based echo sounding, but this method is quite expensive. A relatively new alternative is data obtained by airborne lidar bathymetry. This study investigates the accuracy of these data, which was obtained in the context of the project `Investigation on the use of airborne laser bathymetry in hydrographic surveying'. A comparison to multi-beam echo sounding data shows only small differences in the depths values of the data sets. The IHO requirements of the total horizontal and vertical uncertainty for laser data are met. The second goal of this paper is to compare three spatial interpolation methods, namely Inverse Distance Weighting (IDW), Delaunay Triangulation (TIN), and supervised Artificial Neural Networks (ANN), for the generation of sea bed models. The focus of our investigation is on the amount of required sampling points. This is analyzed by manually reducing the data sets. We found that the three techniques have a similar performance almost independently of the amount of sampling data in our test area. However, ANN are more stable when using a very small subset of points.

NetVLAD: CNN Architecture for Weakly Supervised Place Recognition.

PubMed

Arandjelovic, Relja; Gronat, Petr; Torii, Akihiko; Pajdla, Tomas; Sivic, Josef

2018-06-01

We tackle the problem of large scale visual place recognition, where the task is to quickly and accurately recognize the location of a given query photograph. We present the following four principal contributions. First, we develop a convolutional neural network (CNN) architecture that is trainable in an end-to-end manner directly for the place recognition task. The main component of this architecture, NetVLAD, is a new generalized VLAD layer, inspired by the "Vector of Locally Aggregated Descriptors" image representation commonly used in image retrieval. The layer is readily pluggable into any CNN architecture and amenable to training via backpropagation. Second, we create a new weakly supervised ranking loss, which enables end-to-end learning of the architecture's parameters from images depicting the same places over time downloaded from Google Street View Time Machine. Third, we develop an efficient training procedure which can be applied on very large-scale weakly labelled tasks. Finally, we show that the proposed architecture and training procedure significantly outperform non-learnt image representations and off-the-shelf CNN descriptors on challenging place recognition and image retrieval benchmarks.

Modeling EEG Waveforms with Semi-Supervised Deep Belief Nets: Fast Classification and Anomaly Measurement

PubMed Central

Wulsin, D. F.; Gupta, J. R.; Mani, R.; Blanco, J. A.; Litt, B.

2011-01-01

Clinical electroencephalography (EEG) records vast amounts of human complex data yet is still reviewed primarily by human readers. Deep Belief Nets (DBNs) are a relatively new type of multi-layer neural network commonly tested on two-dimensional image data, but are rarely applied to times-series data such as EEG. We apply DBNs in a semi-supervised paradigm to model EEG waveforms for classification and anomaly detection. DBN performance was comparable to standard classifiers on our EEG dataset, and classification time was found to be 1.7 to 103.7 times faster than the other high-performing classifiers. We demonstrate how the unsupervised step of DBN learning produces an autoencoder that can naturally be used in anomaly measurement. We compare the use of raw, unprocessed data—a rarity in automated physiological waveform analysis—to hand-chosen features and find that raw data produces comparable classification and better anomaly measurement performance. These results indicate that DBNs and raw data inputs may be more effective for online automated EEG waveform recognition than other common techniques. PMID:21525569

Machine learning for outcome prediction of acute ischemic stroke post intra-arterial therapy.

PubMed

Asadi, Hamed; Dowling, Richard; Yan, Bernard; Mitchell, Peter

2014-01-01

Stroke is a major cause of death and disability. Accurately predicting stroke outcome from a set of predictive variables may identify high-risk patients and guide treatment approaches, leading to decreased morbidity. Logistic regression models allow for the identification and validation of predictive variables. However, advanced machine learning algorithms offer an alternative, in particular, for large-scale multi-institutional data, with the advantage of easily incorporating newly available data to improve prediction performance. Our aim was to design and compare different machine learning methods, capable of predicting the outcome of endovascular intervention in acute anterior circulation ischaemic stroke. We conducted a retrospective study of a prospectively collected database of acute ischaemic stroke treated by endovascular intervention. Using SPSS®, MATLAB®, and Rapidminer®, classical statistics as well as artificial neural network and support vector algorithms were applied to design a supervised machine capable of classifying these predictors into potential good and poor outcomes. These algorithms were trained, validated and tested using randomly divided data. We included 107 consecutive acute anterior circulation ischaemic stroke patients treated by endovascular technique. Sixty-six were male and the mean age of 65.3. All the available demographic, procedural and clinical factors were included into the models. The final confusion matrix of the neural network, demonstrated an overall congruency of ∼ 80% between the target and output classes, with favourable receiving operative characteristics. However, after optimisation, the support vector machine had a relatively better performance, with a root mean squared error of 2.064 (SD: ± 0.408). We showed promising accuracy of outcome prediction, using supervised machine learning algorithms, with potential for incorporation of larger multicenter datasets, likely further improving prediction. Finally, we propose that a robust machine learning system can potentially optimise the selection process for endovascular versus medical treatment in the management of acute stroke.

Application of Deep Learning and Supervised Learning Methods to Recognize Nonlinear Hidden Pattern in Water Stress Levels from Spatiotemporal Datasets across Rural and Urban US Counties

NASA Astrophysics Data System (ADS)

Eisenhart, T.; Josset, L.; Rising, J. A.; Devineni, N.; Lall, U.

2017-12-01

In the wake of recent water crises, the need to understand and predict the risk of water stress in urban and rural areas has grown. This understanding has the potential to improve decision making in public resource management, policy making, risk management and investment decisions. Assuming an underlying relationship between urban and rural water stress and observable features, we apply Deep Learning and Supervised Learning models to uncover hidden nonlinear patterns from spatiotemporal datasets. Results of interest includes prediction accuracy on extreme categories (i.e. urban areas highly prone to water stress) and not solely the average risk for urban or rural area, which adds complexity to the tuning of model parameters. We first label urban water stressed counties using annual water quality violations and compile a comprehensive spatiotemporal dataset that captures the yearly evolution of climatic, demographic and economic factors of more than 3,000 US counties over the 1980-2010 period. As county-level data reporting is not done on a yearly basis, we test multiple imputation methods to get around the issue of missing data. Using Python libraries, TensorFlow and scikit-learn, we apply and compare the ability of, amongst other methods, Recurrent Neural Networks (testing both LSTM and GRU cells), Convolutional Neural Networks and Support Vector Machines to predict urban water stress. We evaluate the performance of those models over multiple time spans and combine methods to diminish the risk of overfitting and increase prediction power on test sets. This methodology seeks to identify hidden nonlinear patterns to assess the predominant data features that influence urban and rural water stress. Results from this application at the national scale will assess the performance of deep learning models to predict water stress risk areas across all US counties and will highlight a predominant Machine Learning method for modeling water stress risk using spatiotemporal data.

A supervised 'lesion-enhancement' filter by use of a massive-training artificial neural network (MTANN) in computer-aided diagnosis (CAD).

PubMed

Suzuki, Kenji

2009-09-21

Computer-aided diagnosis (CAD) has been an active area of study in medical image analysis. A filter for the enhancement of lesions plays an important role for improving the sensitivity and specificity in CAD schemes. The filter enhances objects similar to a model employed in the filter; e.g. a blob-enhancement filter based on the Hessian matrix enhances sphere-like objects. Actual lesions, however, often differ from a simple model; e.g. a lung nodule is generally modeled as a solid sphere, but there are nodules of various shapes and with internal inhomogeneities such as a nodule with spiculations and ground-glass opacity. Thus, conventional filters often fail to enhance actual lesions. Our purpose in this study was to develop a supervised filter for the enhancement of actual lesions (as opposed to a lesion model) by use of a massive-training artificial neural network (MTANN) in a CAD scheme for detection of lung nodules in CT. The MTANN filter was trained with actual nodules in CT images to enhance actual patterns of nodules. By use of the MTANN filter, the sensitivity and specificity of our CAD scheme were improved substantially. With a database of 69 lung cancers, nodule candidate detection by the MTANN filter achieved a 97% sensitivity with 6.7 false positives (FPs) per section, whereas nodule candidate detection by a difference-image technique achieved a 96% sensitivity with 19.3 FPs per section. Classification-MTANNs were applied for further reduction of the FPs. The classification-MTANNs removed 60% of the FPs with a loss of one true positive; thus, it achieved a 96% sensitivity with 2.7 FPs per section. Overall, with our CAD scheme based on the MTANN filter and classification-MTANNs, an 84% sensitivity with 0.5 FPs per section was achieved.

Unsupervised learning in persistent sensing for target recognition by wireless ad hoc networks of ground-based sensors

NASA Astrophysics Data System (ADS)

Hortos, William S.

2008-04-01

In previous work by the author, effective persistent and pervasive sensing for recognition and tracking of battlefield targets were seen to be achieved, using intelligent algorithms implemented by distributed mobile agents over a composite system of unmanned aerial vehicles (UAVs) for persistence and a wireless network of unattended ground sensors for pervasive coverage of the mission environment. While simulated performance results for the supervised algorithms of the composite system are shown to provide satisfactory target recognition over relatively brief periods of system operation, this performance can degrade by as much as 50% as target dynamics in the environment evolve beyond the period of system operation in which the training data are representative. To overcome this limitation, this paper applies the distributed approach using mobile agents to the network of ground-based wireless sensors alone, without the UAV subsystem, to provide persistent as well as pervasive sensing for target recognition and tracking. The supervised algorithms used in the earlier work are supplanted by unsupervised routines, including competitive-learning neural networks (CLNNs) and new versions of support vector machines (SVMs) for characterization of an unknown target environment. To capture the same physical phenomena from battlefield targets as the composite system, the suite of ground-based sensors can be expanded to include imaging and video capabilities. The spatial density of deployed sensor nodes is increased to allow more precise ground-based location and tracking of detected targets by active nodes. The "swarm" mobile agents enabling WSN intelligence are organized in a three processing stages: detection, recognition and sustained tracking of ground targets. Features formed from the compressed sensor data are down-selected according to an information-theoretic algorithm that reduces redundancy within the feature set, reducing the dimension of samples used in the target recognition and tracking routines. Target tracking is based on simplified versions of Kalman filtration. Accuracy of recognition and tracking of implemented versions of the proposed suite of unsupervised algorithms is somewhat degraded from the ideal. Target recognition and tracking by supervised routines and by unsupervised SVM and CLNN routines in the ground-based WSN is evaluated in simulations using published system values and sensor data from vehicular targets in ground-surveillance scenarios. Results are compared with previously published performance for the system of the ground-based sensor network (GSN) and UAV swarm.

High resolution depth reconstruction from monocular images and sparse point clouds using deep convolutional neural network

NASA Astrophysics Data System (ADS)

Dimitrievski, Martin; Goossens, Bart; Veelaert, Peter; Philips, Wilfried

2017-09-01

Understanding the 3D structure of the environment is advantageous for many tasks in the field of robotics and autonomous vehicles. From the robot's point of view, 3D perception is often formulated as a depth image reconstruction problem. In the literature, dense depth images are often recovered deterministically from stereo image disparities. Other systems use an expensive LiDAR sensor to produce accurate, but semi-sparse depth images. With the advent of deep learning there have also been attempts to estimate depth by only using monocular images. In this paper we combine the best of the two worlds, focusing on a combination of monocular images and low cost LiDAR point clouds. We explore the idea that very sparse depth information accurately captures the global scene structure while variations in image patches can be used to reconstruct local depth to a high resolution. The main contribution of this paper is a supervised learning depth reconstruction system based on a deep convolutional neural network. The network is trained on RGB image patches reinforced with sparse depth information and the output is a depth estimate for each pixel. Using image and point cloud data from the KITTI vision dataset we are able to learn a correspondence between local RGB information and local depth, while at the same time preserving the global scene structure. Our results are evaluated on sequences from the KITTI dataset and our own recordings using a low cost camera and LiDAR setup.

Selection of neural network structure for system error correction of electro-optical tracker system with horizontal gimbal

NASA Astrophysics Data System (ADS)

Liu, Xing-fa; Cen, Ming

2007-12-01

Neural Network system error correction method is more precise than lest square system error correction method and spheric harmonics function system error correction method. The accuracy of neural network system error correction method is mainly related to the frame of Neural Network. Analysis and simulation prove that both BP neural network system error correction method and RBF neural network system error correction method have high correction accuracy; it is better to use RBF Network system error correction method than BP Network system error correction method for little studying stylebook considering training rate and neural network scale.

Genetic Classification of Populations Using Supervised Learning

PubMed Central

Bridges, Michael; Heron, Elizabeth A.; O'Dushlaine, Colm; Segurado, Ricardo; Morris, Derek; Corvin, Aiden; Gill, Michael; Pinto, Carlos

2011-01-01

There are many instances in genetics in which we wish to determine whether two candidate populations are distinguishable on the basis of their genetic structure. Examples include populations which are geographically separated, case–control studies and quality control (when participants in a study have been genotyped at different laboratories). This latter application is of particular importance in the era of large scale genome wide association studies, when collections of individuals genotyped at different locations are being merged to provide increased power. The traditional method for detecting structure within a population is some form of exploratory technique such as principal components analysis. Such methods, which do not utilise our prior knowledge of the membership of the candidate populations. are termed unsupervised. Supervised methods, on the other hand are able to utilise this prior knowledge when it is available. In this paper we demonstrate that in such cases modern supervised approaches are a more appropriate tool for detecting genetic differences between populations. We apply two such methods, (neural networks and support vector machines) to the classification of three populations (two from Scotland and one from Bulgaria). The sensitivity exhibited by both these methods is considerably higher than that attained by principal components analysis and in fact comfortably exceeds a recently conjectured theoretical limit on the sensitivity of unsupervised methods. In particular, our methods can distinguish between the two Scottish populations, where principal components analysis cannot. We suggest, on the basis of our results that a supervised learning approach should be the method of choice when classifying individuals into pre-defined populations, particularly in quality control for large scale genome wide association studies. PMID:21589856

A novel recurrent neural network with finite-time convergence for linear programming.

PubMed

Liu, Qingshan; Cao, Jinde; Chen, Guanrong

2010-11-01

In this letter, a novel recurrent neural network based on the gradient method is proposed for solving linear programming problems. Finite-time convergence of the proposed neural network is proved by using the Lyapunov method. Compared with the existing neural networks for linear programming, the proposed neural network is globally convergent to exact optimal solutions in finite time, which is remarkable and rare in the literature of neural networks for optimization. Some numerical examples are given to show the effectiveness and excellent performance of the new recurrent neural network.

In-situ trainable intrusion detection system

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

Symons, Christopher T.; Beaver, Justin M.; Gillen, Rob

A computer implemented method detects intrusions using a computer by analyzing network traffic. The method includes a semi-supervised learning module connected to a network node. The learning module uses labeled and unlabeled data to train a semi-supervised machine learning sensor. The method records events that include a feature set made up of unauthorized intrusions and benign computer requests. The method identifies at least some of the benign computer requests that occur during the recording of the events while treating the remainder of the data as unlabeled. The method trains the semi-supervised learning module at the network node in-situ, such thatmore » the semi-supervised learning modules may identify malicious traffic without relying on specific rules, signatures, or anomaly detection.« less

Voxel-Based Neighborhood for Spatial Shape Pattern Classification of Lidar Point Clouds with Supervised Learning.

PubMed

Plaza-Leiva, Victoria; Gomez-Ruiz, Jose Antonio; Mandow, Anthony; García-Cerezo, Alfonso

2017-03-15

Improving the effectiveness of spatial shape features classification from 3D lidar data is very relevant because it is largely used as a fundamental step towards higher level scene understanding challenges of autonomous vehicles and terrestrial robots. In this sense, computing neighborhood for points in dense scans becomes a costly process for both training and classification. This paper proposes a new general framework for implementing and comparing different supervised learning classifiers with a simple voxel-based neighborhood computation where points in each non-overlapping voxel in a regular grid are assigned to the same class by considering features within a support region defined by the voxel itself. The contribution provides offline training and online classification procedures as well as five alternative feature vector definitions based on principal component analysis for scatter, tubular and planar shapes. Moreover, the feasibility of this approach is evaluated by implementing a neural network (NN) method previously proposed by the authors as well as three other supervised learning classifiers found in scene processing methods: support vector machines (SVM), Gaussian processes (GP), and Gaussian mixture models (GMM). A comparative performance analysis is presented using real point clouds from both natural and urban environments and two different 3D rangefinders (a tilting Hokuyo UTM-30LX and a Riegl). Classification performance metrics and processing time measurements confirm the benefits of the NN classifier and the feasibility of voxel-based neighborhood.

Comparison between Two Linear Supervised Learning Machines' Methods with Principle Component Based Methods for the Spectrofluorimetric Determination of Agomelatine and Its Degradants.

PubMed

Elkhoudary, Mahmoud M; Naguib, Ibrahim A; Abdel Salam, Randa A; Hadad, Ghada M

2017-05-01

Four accurate, sensitive and reliable stability indicating chemometric methods were developed for the quantitative determination of Agomelatine (AGM) whether in pure form or in pharmaceutical formulations. Two supervised learning machines' methods; linear artificial neural networks (PC-linANN) preceded by principle component analysis and linear support vector regression (linSVR), were compared with two principle component based methods; principle component regression (PCR) as well as partial least squares (PLS) for the spectrofluorimetric determination of AGM and its degradants. The results showed the benefits behind using linear learning machines' methods and the inherent merits of their algorithms in handling overlapped noisy spectral data especially during the challenging determination of AGM alkaline and acidic degradants (DG1 and DG2). Relative mean squared error of prediction (RMSEP) for the proposed models in the determination of AGM were 1.68, 1.72, 0.68 and 0.22 for PCR, PLS, SVR and PC-linANN; respectively. The results showed the superiority of supervised learning machines' methods over principle component based methods. Besides, the results suggested that linANN is the method of choice for determination of components in low amounts with similar overlapped spectra and narrow linearity range. Comparison between the proposed chemometric models and a reported HPLC method revealed the comparable performance and quantification power of the proposed models.

Neural network models for spatial data mining, map production, and cortical direction selectivity

NASA Astrophysics Data System (ADS)

Parsons, Olga

A family of ARTMAP neural networks for incremental supervised learning has been developed over the last decade. The Sensor Exploitation Group of MIT Lincoln Laboratory (LL) has incorporated an early version of this network as the recognition engine of a hierarchical system for fusion and data mining of multiple registered geospatial images. The LL system has been successfully fielded, but it is limited to target vs. non-target identifications and does not produce whole maps. This dissertation expands the capabilities of the LL system so that it learns to identify arbitrarily many target classes at once and can thus produce a whole map. This new spatial data mining system is designed particularly to cope with the highly skewed class distributions of typical mapping problems. Specification of a consistent procedure and a benchmark testbed has permitted the evaluation of candidate recognition networks as well as pre- and post-processing and feature extraction options. The resulting default ARTMAP network and mapping methodology set a standard for a variety of related mapping problems and application domains. The second part of the dissertation investigates the development of cortical direction selectivity. The possible role of visual experience and oculomotor behavior in the maturation of cells in the primary visual cortex is studied. The responses of neurons in the thalamus and cortex of the cat are modeled when natural scenes are scanned by several types of eye movements. Inspired by the Hebbian-like synaptic plasticity, which is based upon correlations between cell activations, the second-order statistical structure of thalamo-cortical activity is examined. In the simulations, patterns of neural activity that lead to a correct refinement of cell responses are observed during visual fixation, when small ocular movements occur, but are not observed in the presence of large saccades. Simulations also replicate experiments in which kittens are reared under stroboscopic illumination. The abnormal fixational eye movements of these cats may account for the puzzling finding of a specific loss of cortical direction selectivity but preservation of orientation selectivity. This work indicates that the oculomotor behavior of visual fixation may play an important role in the refinement of cell response selectivity.

Linear matrix inequality approach to exponential synchronization of a class of chaotic neural networks with time-varying delays

NASA Astrophysics Data System (ADS)

Wu, Wei; Cui, Bao-Tong

2007-07-01

In this paper, a synchronization scheme for a class of chaotic neural networks with time-varying delays is presented. This class of chaotic neural networks covers several well-known neural networks, such as Hopfield neural networks, cellular neural networks, and bidirectional associative memory networks. The obtained criteria are expressed in terms of linear matrix inequalities, thus they can be efficiently verified. A comparison between our results and the previous results shows that our results are less restrictive.

Recursive least-squares learning algorithms for neural networks

NASA Astrophysics Data System (ADS)

Lewis, Paul S.; Hwang, Jenq N.

1990-11-01

This paper presents the development of a pair of recursive least squares (ItLS) algorithms for online training of multilayer perceptrons which are a class of feedforward artificial neural networks. These algorithms incorporate second order information about the training error surface in order to achieve faster learning rates than are possible using first order gradient descent algorithms such as the generalized delta rule. A least squares formulation is derived from a linearization of the training error function. Individual training pattern errors are linearized about the network parameters that were in effect when the pattern was presented. This permits the recursive solution of the least squares approximation either via conventional RLS recursions or by recursive QR decomposition-based techniques. The computational complexity of the update is 0(N2) where N is the number of network parameters. This is due to the estimation of the N x N inverse Hessian matrix. Less computationally intensive approximations of the ilLS algorithms can be easily derived by using only block diagonal elements of this matrix thereby partitioning the learning into independent sets. A simulation example is presented in which a neural network is trained to approximate a two dimensional Gaussian bump. In this example RLS training required an order of magnitude fewer iterations on average (527) than did training with the generalized delta rule (6 1 BACKGROUND Artificial neural networks (ANNs) offer an interesting and potentially useful paradigm for signal processing and pattern recognition. The majority of ANN applications employ the feed-forward multilayer perceptron (MLP) network architecture in which network parameters are " trained" by a supervised learning algorithm employing the generalized delta rule (GDIt) [1 2]. The GDR algorithm approximates a fixed step steepest descent algorithm using derivatives computed by error backpropagatiori. The GDII algorithm is sometimes referred to as the backpropagation algorithm. However in this paper we will use the term backpropagation to refer only to the process of computing error derivatives. While multilayer perceptrons provide a very powerful nonlinear modeling capability GDR training can be very slow and inefficient. In linear adaptive filtering the analog of the GDR algorithm is the leastmean- squares (LMS) algorithm. Steepest descent-based algorithms such as GDR or LMS are first order because they use only first derivative or gradient information about the training error to be minimized. To speed up the training process second order algorithms may be employed that take advantage of second derivative or Hessian matrix information. Second order information can be incorporated into MLP training in different ways. In many applications especially in the area of pattern recognition the training set is finite. In these cases block learning can be applied using standard nonlinear optimization techniques [3 4 5].

Electronic Neural Networks

NASA Technical Reports Server (NTRS)

Thakoor, Anil

1990-01-01

Viewgraphs on electronic neural networks for space station are presented. Topics covered include: electronic neural networks; electronic implementations; VLSI/thin film hybrid hardware for neurocomputing; computations with analog parallel processing; features of neuroprocessors; applications of neuroprocessors; neural network hardware for terrain trafficability determination; a dedicated processor for path planning; neural network system interface; neural network for robotic control; error backpropagation algorithm for learning; resource allocation matrix; global optimization neuroprocessor; and electrically programmable read only thin-film synaptic array.

The neural network to determine the mechanical properties of the steels

NASA Astrophysics Data System (ADS)

Yemelyanov, Vitaliy; Yemelyanova, Nataliya; Safonova, Marina; Nedelkin, Aleksey

2018-04-01

The authors describe the neural network structure and software that is designed and developed to determine the mechanical properties of steels. The neural network is developed to refine upon the values of the steels properties. The results of simulations of the developed neural network are shown. The authors note the low standard error of the proposed neural network. To realize the proposed neural network the specialized software has been developed.

Differentiating between bipolar and unipolar depression in functional and structural MRI studies.

PubMed

Han, Kyu-Man; De Berardis, Domenico; Fornaro, Michele; Kim, Yong-Ku

2018-03-28

Distinguishing depression in bipolar disorder (BD) from unipolar depression (UD) solely based on clinical clues is difficult, which has led to the exploration of promising neural markers in neuroimaging measures for discriminating between BD depression and UD. In this article, we review structural and functional magnetic resonance imaging (MRI) studies that directly compare UD and BD depression based on neuroimaging modalities including functional MRI studies on regional brain activation or functional connectivity, structural MRI on gray or white matter morphology, and pattern classification analyses using a machine learning approach. Numerous studies have reported distinct functional and structural alterations in emotion- or reward-processing neural circuits between BD depression and UD. Different activation patterns in neural networks including the amygdala, anterior cingulate cortex (ACC), prefrontal cortex (PFC), and striatum during emotion-, reward-, or cognition-related tasks have been reported between BD and UD. A stronger functional connectivity pattern in BD was pronounced in default mode and in frontoparietal networks and brain regions including the PFC, ACC, parietal and temporal regions, and thalamus compared to UD. Gray matter volume differences in the ACC, hippocampus, amygdala, and dorsolateral prefrontal cortex (DLPFC) have been reported between BD and UD, along with a thinner DLPFC in BD compared to UD. BD showed reduced integrity in the anterior part of the corpus callosum and posterior cingulum compared to UD. Several studies performed pattern classification analysis using structural and functional MRI data to distinguish between UD and BD depression using a supervised machine learning approach, which yielded a moderate level of accuracy in classification. Copyright © 2018 Elsevier Inc. All rights reserved.

Source localization in an ocean waveguide using supervised machine learning.

PubMed

Niu, Haiqiang; Reeves, Emma; Gerstoft, Peter

2017-09-01

Source localization in ocean acoustics is posed as a machine learning problem in which data-driven methods learn source ranges directly from observed acoustic data. The pressure received by a vertical linear array is preprocessed by constructing a normalized sample covariance matrix and used as the input for three machine learning methods: feed-forward neural networks (FNN), support vector machines (SVM), and random forests (RF). The range estimation problem is solved both as a classification problem and as a regression problem by these three machine learning algorithms. The results of range estimation for the Noise09 experiment are compared for FNN, SVM, RF, and conventional matched-field processing and demonstrate the potential of machine learning for underwater source localization.

Detection of buried objects by fusing dual-band infrared images

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

Clark, G.A.; Sengupta, S.K.; Sherwood, R.J.

1993-11-01

We have conducted experiments to demonstrate the enhanced detectability of buried land mines using sensor fusion techniques. Multiple sensors, including visible imagery, infrared imagery, and ground penetrating radar (GPR), have been used to acquire data on a number of buried mines and mine surrogates. Because the visible wavelength and GPR data are currently incomplete. This paper focuses on the fusion of two-band infrared images. We use feature-level fusion and supervised learning with the probabilistic neural network (PNN) to evaluate detection performance. The novelty of the work lies in the application of advanced target recognition algorithms, the fusion of dual-band infraredmore » images and evaluation of the techniques using two real data sets.« less

Region stability analysis and tracking control of memristive recurrent neural network.

PubMed

Bao, Gang; Zeng, Zhigang; Shen, Yanjun

2018-02-01

Memristor is firstly postulated by Leon Chua and realized by Hewlett-Packard (HP) laboratory. Research results show that memristor can be used to simulate the synapses of neurons. This paper presents a class of recurrent neural network with HP memristors. Firstly, it shows that memristive recurrent neural network has more compound dynamics than the traditional recurrent neural network by simulations. Then it derives that n dimensional memristive recurrent neural network is composed of [Formula: see text] sub neural networks which do not have a common equilibrium point. By designing the tracking controller, it can make memristive neural network being convergent to the desired sub neural network. At last, two numerical examples are given to verify the validity of our result. Copyright © 2017 Elsevier Ltd. All rights reserved.

A recurrent neural network for nonlinear optimization with a continuously differentiable objective function and bound constraints.

PubMed

Liang, X B; Wang, J

2000-01-01

This paper presents a continuous-time recurrent neural-network model for nonlinear optimization with any continuously differentiable objective function and bound constraints. Quadratic optimization with bound constraints is a special problem which can be solved by the recurrent neural network. The proposed recurrent neural network has the following characteristics. 1) It is regular in the sense that any optimum of the objective function with bound constraints is also an equilibrium point of the neural network. If the objective function to be minimized is convex, then the recurrent neural network is complete in the sense that the set of optima of the function with bound constraints coincides with the set of equilibria of the neural network. 2) The recurrent neural network is primal and quasiconvergent in the sense that its trajectory cannot escape from the feasible region and will converge to the set of equilibria of the neural network for any initial point in the feasible bound region. 3) The recurrent neural network has an attractivity property in the sense that its trajectory will eventually converge to the feasible region for any initial states even at outside of the bounded feasible region. 4) For minimizing any strictly convex quadratic objective function subject to bound constraints, the recurrent neural network is globally exponentially stable for almost any positive network parameters. Simulation results are given to demonstrate the convergence and performance of the proposed recurrent neural network for nonlinear optimization with bound constraints.

Neural network-based nonlinear model predictive control vs. linear quadratic gaussian control

USGS Publications Warehouse

Cho, C.; Vance, R.; Mardi, N.; Qian, Z.; Prisbrey, K.

1997-01-01

One problem with the application of neural networks to the multivariable control of mineral and extractive processes is determining whether and how to use them. The objective of this investigation was to compare neural network control to more conventional strategies and to determine if there are any advantages in using neural network control in terms of set-point tracking, rise time, settling time, disturbance rejection and other criteria. The procedure involved developing neural network controllers using both historical plant data and simulation models. Various control patterns were tried, including both inverse and direct neural network plant models. These were compared to state space controllers that are, by nature, linear. For grinding and leaching circuits, a nonlinear neural network-based model predictive control strategy was superior to a state space-based linear quadratic gaussian controller. The investigation pointed out the importance of incorporating state space into neural networks by making them recurrent, i.e., feeding certain output state variables into input nodes in the neural network. It was concluded that neural network controllers can have better disturbance rejection, set-point tracking, rise time, settling time and lower set-point overshoot, and it was also concluded that neural network controllers can be more reliable and easy to implement in complex, multivariable plants.

An Introduction to Neural Networks for Hearing Aid Noise Recognition.

ERIC Educational Resources Information Center

Kim, Jun W.; Tyler, Richard S.

1995-01-01

This article introduces the use of multilayered artificial neural networks in hearing aid noise recognition. It reviews basic principles of neural networks, and offers an example of an application in which a neural network is used to identify the presence or absence of noise in speech. The ability of neural networks to "learn" the…

Application of neural nets in structural optimization

NASA Technical Reports Server (NTRS)

Berke, Laszlo; Hajela, Prabhat

1993-01-01

The biological motivation for Artificial Neural Net developments is briefly discussed, and the most popular paradigm, the feedforward supervised learning net with error back propagation training algorithm, is introduced. Possible approaches for utilization in structural optimization is illustrated through simple examples. Other currently ongoing developments for application in structural mechanics are also mentioned.

Relieving the Bottleneck: An Investigation of Barriers to Expansion of Supervision Networks at Genetic Counseling Training Programs.

PubMed

Berg, Jordan; Hoskovec, Jennifer; Hashmi, S Shahrukh; McCarthy Veach, Patricia; Ownby, Allison; Singletary, Claire N

2018-02-01

Rapid growth in the demand for genetic counselors has led to a workforce shortage. There is a prevailing assumption that the number of training slots for genetic counseling students is linked to the availability of clinical supervisors. This study aimed to determine and compare barriers to expansion of supervision networks at genetic counseling training programs as perceived by supervisors, non-supervisors, and Program Directors. Genetic counselors were recruited via National Society of Genetic Counselors e-blast; Program Directors received personal emails. Online surveys were completed by 216 supervisors, 98 non-supervisors, and 23 Program Directors. Respondents rated impact of 35 barriers; comparisons were made using Kruskal-Wallis and Wilcoxon ranked sum tests. Half of supervisors (51%) indicated willingness to increase supervision. All non-supervisors were willing to supervise. However, all agreed that being too busy impacted ability to supervise, highlighted by supervisors' most impactful barriers: lack of time, other responsibilities, intensive nature of supervision, desire for breaks, and unfilled positions. Non-supervisors noted unique barriers: distance, institutional barriers, and non-clinical roles. Program Directors' perceptions were congruent with those of genetic counselors with three exceptions they rated as impactful: lack of money, prefer not to supervise, and never been asked. In order to expand supervision networks and provide comprehensive student experiences, the profession must examine service delivery models to increase workplace efficiency, reconsider the supervision paradigm, and redefine what constitutes a countable case or place value on non-direct patient care experiences.

Quantized Synchronization of Chaotic Neural Networks With Scheduled Output Feedback Control.

PubMed

Wan, Ying; Cao, Jinde; Wen, Guanghui

In this paper, the synchronization problem of master-slave chaotic neural networks with remote sensors, quantization process, and communication time delays is investigated. The information communication channel between the master chaotic neural network and slave chaotic neural network consists of several remote sensors, with each sensor able to access only partial knowledge of output information of the master neural network. At each sampling instants, each sensor updates its own measurement and only one sensor is scheduled to transmit its latest information to the controller's side in order to update the control inputs for the slave neural network. Thus, such communication process and control strategy are much more energy-saving comparing with the traditional point-to-point scheme. Sufficient conditions for output feedback control gain matrix, allowable length of sampling intervals, and upper bound of network-induced delays are derived to ensure the quantized synchronization of master-slave chaotic neural networks. Lastly, Chua's circuit system and 4-D Hopfield neural network are simulated to validate the effectiveness of the main results.In this paper, the synchronization problem of master-slave chaotic neural networks with remote sensors, quantization process, and communication time delays is investigated. The information communication channel between the master chaotic neural network and slave chaotic neural network consists of several remote sensors, with each sensor able to access only partial knowledge of output information of the master neural network. At each sampling instants, each sensor updates its own measurement and only one sensor is scheduled to transmit its latest information to the controller's side in order to update the control inputs for the slave neural network. Thus, such communication process and control strategy are much more energy-saving comparing with the traditional point-to-point scheme. Sufficient conditions for output feedback control gain matrix, allowable length of sampling intervals, and upper bound of network-induced delays are derived to ensure the quantized synchronization of master-slave chaotic neural networks. Lastly, Chua's circuit system and 4-D Hopfield neural network are simulated to validate the effectiveness of the main results.

Mentoring, Networking and Supervision: Parallelogram, Vortex, or Merging Point?

ERIC Educational Resources Information Center

Hernandez, Mary N.

1994-01-01

Discussion of recruiting and hiring minorities in academic libraries focuses on steps needed for the retention of these minorities. Highlights include literature search strategies; mentoring systems; networking within and outside the organization; and supervision, including immediate feedback. (Contains 20 references.) (LRW)

Modified neural networks for rapid recovery of tokamak plasma parameters for real time control

NASA Astrophysics Data System (ADS)

Sengupta, A.; Ranjan, P.

2002-07-01

Two modified neural network techniques are used for the identification of the equilibrium plasma parameters of the Superconducting Steady State Tokamak I from external magnetic measurements. This is expected to ultimately assist in a real time plasma control. As different from the conventional network structure where a single network with the optimum number of processing elements calculates the outputs, a multinetwork system connected in parallel does the calculations here in one of the methods. This network is called the double neural network. The accuracy of the recovered parameters is clearly more than the conventional network. The other type of neural network used here is based on the statistical function parametrization combined with a neural network. The principal component transformation removes linear dependences from the measurements and a dimensional reduction process reduces the dimensionality of the input space. This reduced and transformed input set, rather than the entire set, is fed into the neural network input. This is known as the principal component transformation-based neural network. The accuracy of the recovered parameters in the latter type of modified network is found to be a further improvement over the accuracy of the double neural network. This result differs from that obtained in an earlier work where the double neural network showed better performance. The conventional network and the function parametrization methods have also been used for comparison. The conventional network has been used for an optimization of the set of magnetic diagnostics. The effective set of sensors, as assessed by this network, are compared with the principal component based network. Fault tolerance of the neural networks has been tested. The double neural network showed the maximum resistance to faults in the diagnostics, while the principal component based network performed poorly. Finally the processing times of the methods have been compared. The double network and the principal component network involve the minimum computation time, although the conventional network also performs well enough to be used in real time.

Control of magnetic bearing systems via the Chebyshev polynomial-based unified model (CPBUM) neural network.

PubMed

Jeng, J T; Lee, T T

2000-01-01

A Chebyshev polynomial-based unified model (CPBUM) neural network is introduced and applied to control a magnetic bearing systems. First, we show that the CPBUM neural network not only has the same capability of universal approximator, but also has faster learning speed than conventional feedforward/recurrent neural network. It turns out that the CPBUM neural network is more suitable in the design of controller than the conventional feedforward/recurrent neural network. Second, we propose the inverse system method, based on the CPBUM neural networks, to control a magnetic bearing system. The proposed controller has two structures; namely, off-line and on-line learning structures. We derive a new learning algorithm for each proposed structure. The experimental results show that the proposed neural network architecture provides a greater flexibility and better performance in controlling magnetic bearing systems.

ChainMail based neural dynamics modeling of soft tissue deformation for surgical simulation.

PubMed

Zhang, Jinao; Zhong, Yongmin; Smith, Julian; Gu, Chengfan

2017-07-20

Realistic and real-time modeling and simulation of soft tissue deformation is a fundamental research issue in the field of surgical simulation. In this paper, a novel cellular neural network approach is presented for modeling and simulation of soft tissue deformation by combining neural dynamics of cellular neural network with ChainMail mechanism. The proposed method formulates the problem of elastic deformation into cellular neural network activities to avoid the complex computation of elasticity. The local position adjustments of ChainMail are incorporated into the cellular neural network as the local connectivity of cells, through which the dynamic behaviors of soft tissue deformation are transformed into the neural dynamics of cellular neural network. Experiments demonstrate that the proposed neural network approach is capable of modeling the soft tissues' nonlinear deformation and typical mechanical behaviors. The proposed method not only improves ChainMail's linear deformation with the nonlinear characteristics of neural dynamics but also enables the cellular neural network to follow the principle of continuum mechanics to simulate soft tissue deformation.

Familiarity Detection is an Intrinsic Property of Cortical Microcircuits with Bidirectional Synaptic Plasticity.

PubMed

Zhang, Xiaoyu; Ju, Han; Penney, Trevor B; VanDongen, Antonius M J

2017-01-01

Humans instantly recognize a previously seen face as "familiar." To deepen our understanding of familiarity-novelty detection, we simulated biologically plausible neural network models of generic cortical microcircuits consisting of spiking neurons with random recurrent synaptic connections. NMDA receptor (NMDAR)-dependent synaptic plasticity was implemented to allow for unsupervised learning and bidirectional modifications. Network spiking activity evoked by sensory inputs consisting of face images altered synaptic efficacy, which resulted in the network responding more strongly to a previously seen face than a novel face. Network size determined how many faces could be accurately recognized as familiar. When the simulated model became sufficiently complex in structure, multiple familiarity traces could be retained in the same network by forming partially-overlapping subnetworks that differ slightly from each other, thereby resulting in a high storage capacity. Fisher's discriminant analysis was applied to identify critical neurons whose spiking activity predicted familiar input patterns. Intriguingly, as sensory exposure was prolonged, the selected critical neurons tended to appear at deeper layers of the network model, suggesting recruitment of additional circuits in the network for incremental information storage. We conclude that generic cortical microcircuits with bidirectional synaptic plasticity have an intrinsic ability to detect familiar inputs. This ability does not require a specialized wiring diagram or supervision and can therefore be expected to emerge naturally in developing cortical circuits.

Familiarity Detection is an Intrinsic Property of Cortical Microcircuits with Bidirectional Synaptic Plasticity

PubMed Central

2017-01-01

Abstract Humans instantly recognize a previously seen face as “familiar.” To deepen our understanding of familiarity-novelty detection, we simulated biologically plausible neural network models of generic cortical microcircuits consisting of spiking neurons with random recurrent synaptic connections. NMDA receptor (NMDAR)-dependent synaptic plasticity was implemented to allow for unsupervised learning and bidirectional modifications. Network spiking activity evoked by sensory inputs consisting of face images altered synaptic efficacy, which resulted in the network responding more strongly to a previously seen face than a novel face. Network size determined how many faces could be accurately recognized as familiar. When the simulated model became sufficiently complex in structure, multiple familiarity traces could be retained in the same network by forming partially-overlapping subnetworks that differ slightly from each other, thereby resulting in a high storage capacity. Fisher’s discriminant analysis was applied to identify critical neurons whose spiking activity predicted familiar input patterns. Intriguingly, as sensory exposure was prolonged, the selected critical neurons tended to appear at deeper layers of the network model, suggesting recruitment of additional circuits in the network for incremental information storage. We conclude that generic cortical microcircuits with bidirectional synaptic plasticity have an intrinsic ability to detect familiar inputs. This ability does not require a specialized wiring diagram or supervision and can therefore be expected to emerge naturally in developing cortical circuits. PMID:28534043

Nested Neural Networks

NASA Technical Reports Server (NTRS)

Baram, Yoram

1992-01-01

Report presents analysis of nested neural networks, consisting of interconnected subnetworks. Analysis based on simplified mathematical models more appropriate for artificial electronic neural networks, partly applicable to biological neural networks. Nested structure allows for retrieval of individual subpatterns. Requires fewer wires and connection devices than fully connected networks, and allows for local reconstruction of damaged subnetworks without rewiring entire network.

Quantum neural networks: Current status and prospects for development

NASA Astrophysics Data System (ADS)

Altaisky, M. V.; Kaputkina, N. E.; Krylov, V. A.

2014-11-01

The idea of quantum artificial neural networks, first formulated in [34], unites the artificial neural network concept with the quantum computation paradigm. Quantum artificial neural networks were first systematically considered in the PhD thesis by T. Menneer (1998). Based on the works of Menneer and Narayanan [42, 43], Kouda, Matsui, and Nishimura [35, 36], Altaisky [2, 68], Zhou [67], and others, quantum-inspired learning algorithms for neural networks were developed, and are now used in various training programs and computer games [29, 30]. The first practically realizable scaled hardware-implemented model of the quantum artificial neural network is obtained by D-Wave Systems, Inc. [33]. It is a quantum Hopfield network implemented on the basis of superconducting quantum interference devices (SQUIDs). In this work we analyze possibilities and underlying principles of an alternative way to implement quantum neural networks on the basis of quantum dots. A possibility of using quantum neural network algorithms in automated control systems, associative memory devices, and in modeling biological and social networks is examined.

Neural network approaches to capture temporal information

NASA Astrophysics Data System (ADS)

van Veelen, Martijn; Nijhuis, Jos; Spaanenburg, Ben

2000-05-01

The automated design and construction of neural networks receives growing attention of the neural networks community. Both the growing availability of computing power and development of mathematical and probabilistic theory have had severe impact on the design and modelling approaches of neural networks. This impact is most apparent in the use of neural networks to time series prediction. In this paper, we give our views on past, contemporary and future design and modelling approaches to neural forecasting.

The role of symmetry in neural networks and their Laplacian spectra.

PubMed

de Lange, Siemon C; van den Heuvel, Martijn P; de Reus, Marcel A

2016-11-01

Human and animal nervous systems constitute complexly wired networks that form the infrastructure for neural processing and integration of information. The organization of these neural networks can be analyzed using the so-called Laplacian spectrum, providing a mathematical tool to produce systems-level network fingerprints. In this article, we examine a characteristic central peak in the spectrum of neural networks, including anatomical brain network maps of the mouse, cat and macaque, as well as anatomical and functional network maps of human brain connectivity. We link the occurrence of this central peak to the level of symmetry in neural networks, an intriguing aspect of network organization resulting from network elements that exhibit similar wiring patterns. Specifically, we propose a measure to capture the global level of symmetry of a network and show that, for both empirical networks and network models, the height of the main peak in the Laplacian spectrum is strongly related to node symmetry in the underlying network. Moreover, examination of spectra of duplication-based model networks shows that neural spectra are best approximated using a trade-off between duplication and diversification. Taken together, our results facilitate a better understanding of neural network spectra and the importance of symmetry in neural networks. Copyright © 2016 Elsevier Inc. All rights reserved.

Synchronization Control of Neural Networks With State-Dependent Coefficient Matrices.

PubMed

Zhang, Junfeng; Zhao, Xudong; Huang, Jun

2016-11-01

This brief is concerned with synchronization control of a class of neural networks with state-dependent coefficient matrices. Being different from the existing drive-response neural networks in the literature, a novel model of drive-response neural networks is established. The concepts of uniformly ultimately bounded (UUB) synchronization and convex hull Lyapunov function are introduced. Then, by using the convex hull Lyapunov function approach, the UUB synchronization design of the drive-response neural networks is proposed, and a delay-independent control law guaranteeing the bounded synchronization of the neural networks is constructed. All present conditions are formulated in terms of bilinear matrix inequalities. By comparison, it is shown that the neural networks obtained in this brief are less conservative than those ones in the literature, and the bounded synchronization is suitable for the novel drive-response neural networks. Finally, an illustrative example is given to verify the validity of the obtained results.

The Laplacian spectrum of neural networks

PubMed Central

de Lange, Siemon C.; de Reus, Marcel A.; van den Heuvel, Martijn P.

2014-01-01

The brain is a complex network of neural interactions, both at the microscopic and macroscopic level. Graph theory is well suited to examine the global network architecture of these neural networks. Many popular graph metrics, however, encode average properties of individual network elements. Complementing these “conventional” graph metrics, the eigenvalue spectrum of the normalized Laplacian describes a network's structure directly at a systems level, without referring to individual nodes or connections. In this paper, the Laplacian spectra of the macroscopic anatomical neuronal networks of the macaque and cat, and the microscopic network of the Caenorhabditis elegans were examined. Consistent with conventional graph metrics, analysis of the Laplacian spectra revealed an integrative community structure in neural brain networks. Extending previous findings of overlap of network attributes across species, similarity of the Laplacian spectra across the cat, macaque and C. elegans neural networks suggests a certain level of consistency in the overall architecture of the anatomical neural networks of these species. Our results further suggest a specific network class for neural networks, distinct from conceptual small-world and scale-free models as well as several empirical networks. PMID:24454286

Introduction to Neural Networks.

DTIC Science & Technology

1992-03-01

parallel processing of information that can greatly reduce the time required to perform operations which are needed in pattern recognition. Neural network, Artificial neural network , Neural net, ANN.

Learning control of inverted pendulum system by neural network driven fuzzy reasoning: The learning function of NN-driven fuzzy reasoning under changes of reasoning environment

NASA Technical Reports Server (NTRS)

Hayashi, Isao; Nomura, Hiroyoshi; Wakami, Noboru

1991-01-01

Whereas conventional fuzzy reasonings are associated with tuning problems, which are lack of membership functions and inference rule designs, a neural network driven fuzzy reasoning (NDF) capable of determining membership functions by neural network is formulated. In the antecedent parts of the neural network driven fuzzy reasoning, the optimum membership function is determined by a neural network, while in the consequent parts, an amount of control for each rule is determined by other plural neural networks. By introducing an algorithm of neural network driven fuzzy reasoning, inference rules for making a pendulum stand up from its lowest suspended point are determined for verifying the usefulness of the algorithm.

Optimization of neural network architecture using genetic programming improves detection and modeling of gene-gene interactions in studies of human diseases

PubMed Central

Ritchie, Marylyn D; White, Bill C; Parker, Joel S; Hahn, Lance W; Moore, Jason H

2003-01-01

Background Appropriate definition of neural network architecture prior to data analysis is crucial for successful data mining. This can be challenging when the underlying model of the data is unknown. The goal of this study was to determine whether optimizing neural network architecture using genetic programming as a machine learning strategy would improve the ability of neural networks to model and detect nonlinear interactions among genes in studies of common human diseases. Results Using simulated data, we show that a genetic programming optimized neural network approach is able to model gene-gene interactions as well as a traditional back propagation neural network. Furthermore, the genetic programming optimized neural network is better than the traditional back propagation neural network approach in terms of predictive ability and power to detect gene-gene interactions when non-functional polymorphisms are present. Conclusion This study suggests that a machine learning strategy for optimizing neural network architecture may be preferable to traditional trial-and-error approaches for the identification and characterization of gene-gene interactions in common, complex human diseases. PMID:12846935

Medical image analysis with artificial neural networks.

PubMed

Jiang, J; Trundle, P; Ren, J

2010-12-01

Given that neural networks have been widely reported in the research community of medical imaging, we provide a focused literature survey on recent neural network developments in computer-aided diagnosis, medical image segmentation and edge detection towards visual content analysis, and medical image registration for its pre-processing and post-processing, with the aims of increasing awareness of how neural networks can be applied to these areas and to provide a foundation for further research and practical development. Representative techniques and algorithms are explained in detail to provide inspiring examples illustrating: (i) how a known neural network with fixed structure and training procedure could be applied to resolve a medical imaging problem; (ii) how medical images could be analysed, processed, and characterised by neural networks; and (iii) how neural networks could be expanded further to resolve problems relevant to medical imaging. In the concluding section, a highlight of comparisons among many neural network applications is included to provide a global view on computational intelligence with neural networks in medical imaging. Copyright © 2010 Elsevier Ltd. All rights reserved.

Application of artificial neural network to investigate the effects of 5-fluorouracil on ribonucleotides and deoxyribonucleotides in HepG2 cells

PubMed Central

Guo, Jianru; Chen, QianQian; Lam, Christopher Wai Kei; Wang, Caiyun; Wong, Vincent Kam Wai; Xu, Fengguo; Jiang, ZhiHong; Zhang, Wei

2015-01-01

Endogenous ribonucleotides and deoxyribonucleotides are essential metabolites that play important roles in a broad range of key cellular functions. Their intracellular levels could also reflect the action of nucleoside analogues. We investigated the effects of 5-fluorouracil (5-FU) on ribonucleotide and deoxyribonucleotide pool sizes in cells upon exposure to 5-FU for different durations. Unsupervised and supervised artificial neural networks were compared for comprehensive analysis of global responses to 5-FU. As expected, deoxyuridine monophosphate (dUMP) increased after 5-FU incubation due to the inhibition of thymine monophosphate (TMP) synthesis. Interestingly, the accumulation of dUMP could not lead to increased levels of deoxyuridine triphosphate (dUTP) and deoxyuridine diphosphate (dUDP). After the initial fall in intracellular deoxythymidine triphosphate (TTP) concentration, its level recovered and increased from 48 h exposure to 5-FU, although deoxythymidine diphosphate (TDP) and TMP continued to decrease compared with the control group. These findings suggest 5-FU treatment caused unexpected changes in intracellular purine polls, such as increases in deoxyadenosine triphosphate (dATP), adenosine-triphosphate (ATP), guanosine triphosphate (GTP) pools. Further elucidation of the mechanism of action of 5-FU in causing these changes should enhance development of strategies that will increase the anticancer activity of 5-FU while decreasing its resistance. PMID:26578061

Novel Screening Tool for Stroke Using Artificial Neural Network.

PubMed

Abedi, Vida; Goyal, Nitin; Tsivgoulis, Georgios; Hosseinichimeh, Niyousha; Hontecillas, Raquel; Bassaganya-Riera, Josep; Elijovich, Lucas; Metter, Jeffrey E; Alexandrov, Anne W; Liebeskind, David S; Alexandrov, Andrei V; Zand, Ramin

2017-06-01

The timely diagnosis of stroke at the initial examination is extremely important given the disease morbidity and narrow time window for intervention. The goal of this study was to develop a supervised learning method to recognize acute cerebral ischemia (ACI) and differentiate that from stroke mimics in an emergency setting. Consecutive patients presenting to the emergency department with stroke-like symptoms, within 4.5 hours of symptoms onset, in 2 tertiary care stroke centers were randomized for inclusion in the model. We developed an artificial neural network (ANN) model. The learning algorithm was based on backpropagation. To validate the model, we used a 10-fold cross-validation method. A total of 260 patients (equal number of stroke mimics and ACIs) were enrolled for the development and validation of our ANN model. Our analysis indicated that the average sensitivity and specificity of ANN for the diagnosis of ACI based on the 10-fold cross-validation analysis was 80.0% (95% confidence interval, 71.8-86.3) and 86.2% (95% confidence interval, 78.7-91.4), respectively. The median precision of ANN for the diagnosis of ACI was 92% (95% confidence interval, 88.7-95.3). Our results show that ANN can be an effective tool for the recognition of ACI and differentiation of ACI from stroke mimics at the initial examination. © 2017 American Heart Association, Inc.

Tuberculosis diagnosis support analysis for precarious health information systems.

PubMed

Orjuela-Cañón, Alvaro David; Camargo Mendoza, Jorge Eliécer; Awad García, Carlos Enrique; Vergara Vela, Erika Paola

2018-04-01

Pulmonary tuberculosis is a world emergency for the World Health Organization. Techniques and new diagnosis tools are important to battle this bacterial infection. There have been many advances in all those fields, but in developing countries such as Colombia, where the resources and infrastructure are limited, new fast and less expensive strategies are increasingly needed. Artificial neural networks are computational intelligence techniques that can be used in this kind of problems and offer additional support in the tuberculosis diagnosis process, providing a tool to medical staff to make decisions about management of subjects under suspicious of tuberculosis. A database extracted from 105 subjects with precarious information of people under suspect of pulmonary tuberculosis was used in this study. Data extracted from sex, age, diabetes, homeless, AIDS status and a variable with clinical knowledge from the medical personnel were used. Models based on artificial neural networks were used, exploring supervised learning to detect the disease. Unsupervised learning was used to create three risk groups based on available information. Obtained results are comparable with traditional techniques for detection of tuberculosis, showing advantages such as fast and low implementation costs. Sensitivity of 97% and specificity of 71% where achieved. Used techniques allowed to obtain valuable information that can be useful for physicians who treat the disease in decision making processes, especially under limited infrastructure and data. Copyright © 2018 Elsevier B.V. All rights reserved.

Supervised detection of exoplanets in high-contrast imaging sequences

NASA Astrophysics Data System (ADS)

Gomez Gonzalez, C. A.; Absil, O.; Van Droogenbroeck, M.

2018-06-01

Context. Post-processing algorithms play a key role in pushing the detection limits of high-contrast imaging (HCI) instruments. State-of-the-art image processing approaches for HCI enable the production of science-ready images relying on unsupervised learning techniques, such as low-rank approximations, for generating a model point spread function (PSF) and subtracting the residual starlight and speckle noise. Aims: In order to maximize the detection rate of HCI instruments and survey campaigns, advanced algorithms with higher sensitivities to faint companions are needed, especially for the speckle-dominated innermost region of the images. Methods: We propose a reformulation of the exoplanet detection task (for ADI sequences) that builds on well-established machine learning techniques to take HCI post-processing from an unsupervised to a supervised learning context. In this new framework, we present algorithmic solutions using two different discriminative models: SODIRF (random forests) and SODINN (neural networks). We test these algorithms on real ADI datasets from VLT/NACO and VLT/SPHERE HCI instruments. We then assess their performances by injecting fake companions and using receiver operating characteristic analysis. This is done in comparison with state-of-the-art ADI algorithms, such as ADI principal component analysis (ADI-PCA). Results: This study shows the improved sensitivity versus specificity trade-off of the proposed supervised detection approach. At the diffraction limit, SODINN improves the true positive rate by a factor ranging from 2 to 10 (depending on the dataset and angular separation) with respect to ADI-PCA when working at the same false-positive level. Conclusions: The proposed supervised detection framework outperforms state-of-the-art techniques in the task of discriminating planet signal from speckles. In addition, it offers the possibility of re-processing existing HCI databases to maximize their scientific return and potentially improve the demographics of directly imaged exoplanets.

Neural-network-directed alignment of optical systems using the laser-beam spatial filter as an example

NASA Technical Reports Server (NTRS)

Decker, Arthur J.; Krasowski, Michael J.; Weiland, Kenneth E.

1993-01-01

This report describes an effort at NASA Lewis Research Center to use artificial neural networks to automate the alignment and control of optical measurement systems. Specifically, it addresses the use of commercially available neural network software and hardware to direct alignments of the common laser-beam-smoothing spatial filter. The report presents a general approach for designing alignment records and combining these into training sets to teach optical alignment functions to neural networks and discusses the use of these training sets to train several types of neural networks. Neural network configurations used include the adaptive resonance network, the back-propagation-trained network, and the counter-propagation network. This work shows that neural networks can be used to produce robust sequencers. These sequencers can learn by example to execute the step-by-step procedures of optical alignment and also can learn adaptively to correct for environmentally induced misalignment. The long-range objective is to use neural networks to automate the alignment and operation of optical measurement systems in remote, harsh, or dangerous aerospace environments. This work also shows that when neural networks are trained by a human operator, training sets should be recorded, training should be executed, and testing should be done in a manner that does not depend on intellectual judgments of the human operator.

Resolution of Singularities Introduced by Hierarchical Structure in Deep Neural Networks.

PubMed

Nitta, Tohru

2017-10-01

We present a theoretical analysis of singular points of artificial deep neural networks, resulting in providing deep neural network models having no critical points introduced by a hierarchical structure. It is considered that such deep neural network models have good nature for gradient-based optimization. First, we show that there exist a large number of critical points introduced by a hierarchical structure in deep neural networks as straight lines, depending on the number of hidden layers and the number of hidden neurons. Second, we derive a sufficient condition for deep neural networks having no critical points introduced by a hierarchical structure, which can be applied to general deep neural networks. It is also shown that the existence of critical points introduced by a hierarchical structure is determined by the rank and the regularity of weight matrices for a specific class of deep neural networks. Finally, two kinds of implementation methods of the sufficient conditions to have no critical points are provided. One is a learning algorithm that can avoid critical points introduced by the hierarchical structure during learning (called avoidant learning algorithm). The other is a neural network that does not have some critical points introduced by the hierarchical structure as an inherent property (called avoidant neural network).

The effect of the neural activity on topological properties of growing neural networks.

PubMed

Gafarov, F M; Gafarova, V R

2016-09-01

The connectivity structure in cortical networks defines how information is transmitted and processed, and it is a source of the complex spatiotemporal patterns of network's development, and the process of creation and deletion of connections is continuous in the whole life of the organism. In this paper, we study how neural activity influences the growth process in neural networks. By using a two-dimensional activity-dependent growth model we demonstrated the neural network growth process from disconnected neurons to fully connected networks. For making quantitative investigation of the network's activity influence on its topological properties we compared it with the random growth network not depending on network's activity. By using the random graphs theory methods for the analysis of the network's connections structure it is shown that the growth in neural networks results in the formation of a well-known "small-world" network.

Odor source identification by grounding linguistic descriptions in an artificial nose

NASA Astrophysics Data System (ADS)

Loutfi, Amy; Coradeschi, Silvia; Duckett, Tom; Wide, Peter

2001-03-01

This paper addresses the problem of enabling autonomous agents (e.g., robots) to carry out human oriented tasks using an electronic nose. The nose consists of a combination of passive gas sensors with different selectivity, the outputs of which are fused together with an artificial neural network in order to recognize various human-determined odors. The basic idea is to ground human-provided linguistic descriptions of these odors in the actual sensory perceptions of the nose through a process of supervised learning. Analogous to the human nose, the paper explains a method by which an electronic nose can be used for substance identification. First, the receptors of the nose are exposed to a substance by means of inhalation with an electric pump. Then a chemical reaction takes place in the gas sensors over a period of time and an artificial neural network processes the resulting sensor patterns. This network was trained to recognize a basic set of pure substances such as vanilla, lavender and yogurt under controlled laboratory conditions. The complete system was then validated through a series of experiments on various combinations of the basic substances. First, we showed that the nose was able to consistently recognize unseen samples of the same substances on which it had been trained. In addition, we presented some first results where the nose was tested on novel combinations of substances on which it had not been trained by combining the learned descriptions - for example, it could distinguish lavender yogurt as a combination of lavender and yogurt.

LavaNet—Neural network development environment in a general mine planning package

NASA Astrophysics Data System (ADS)

Kapageridis, Ioannis Konstantinou; Triantafyllou, A. G.

2011-04-01

LavaNet is a series of scripts written in Perl that gives access to a neural network simulation environment inside a general mine planning package. A well known and a very popular neural network development environment, the Stuttgart Neural Network Simulator, is used as the base for the development of neural networks. LavaNet runs inside VULCAN™—a complete mine planning package with advanced database, modelling and visualisation capabilities. LavaNet is taking advantage of VULCAN's Perl based scripting environment, Lava, to bring all the benefits of neural network development and application to geologists, mining engineers and other users of the specific mine planning package. LavaNet enables easy development of neural network training data sets using information from any of the data and model structures available, such as block models and drillhole databases. Neural networks can be trained inside VULCAN™ and the results be used to generate new models that can be visualised in 3D. Direct comparison of developed neural network models with conventional and geostatistical techniques is now possible within the same mine planning software package. LavaNet supports Radial Basis Function networks, Multi-Layer Perceptrons and Self-Organised Maps.

Creative-Dynamics Approach To Neural Intelligence

NASA Technical Reports Server (NTRS)

Zak, Michail A.

1992-01-01

Paper discusses approach to mathematical modeling of artificial neural networks exhibiting complicated behaviors reminiscent of creativity and intelligence of biological neural networks. Neural network treated as non-Lipschitzian dynamical system - as described in "Non-Lipschitzian Dynamics For Modeling Neural Networks" (NPO-17814). System serves as tool for modeling of temporal-pattern memories and recognition of complicated spatial patterns.

An Attractor-Based Complexity Measurement for Boolean Recurrent Neural Networks

PubMed Central

Cabessa, Jérémie; Villa, Alessandro E. P.

2014-01-01

We provide a novel refined attractor-based complexity measurement for Boolean recurrent neural networks that represents an assessment of their computational power in terms of the significance of their attractor dynamics. This complexity measurement is achieved by first proving a computational equivalence between Boolean recurrent neural networks and some specific class of -automata, and then translating the most refined classification of -automata to the Boolean neural network context. As a result, a hierarchical classification of Boolean neural networks based on their attractive dynamics is obtained, thus providing a novel refined attractor-based complexity measurement for Boolean recurrent neural networks. These results provide new theoretical insights to the computational and dynamical capabilities of neural networks according to their attractive potentialities. An application of our findings is illustrated by the analysis of the dynamics of a simplified model of the basal ganglia-thalamocortical network simulated by a Boolean recurrent neural network. This example shows the significance of measuring network complexity, and how our results bear new founding elements for the understanding of the complexity of real brain circuits. PMID:24727866

Groundwater vulnerability indices conditioned by Supervised Intelligence Committee Machine (SICM).

PubMed

Nadiri, Ata Allah; Gharekhani, Maryam; Khatibi, Rahman; Sadeghfam, Sina; Moghaddam, Asghar Asghari

2017-01-01

This research presents a Supervised Intelligent Committee Machine (SICM) model to assess groundwater vulnerability indices of an aquifer. SICM uses Artificial Neural Networks (ANN) to overarch three Artificial Intelligence (AI) models: Support Vector Machine (SVM), Neuro-Fuzzy (NF) and Gene Expression Programming (GEP). Each model uses the DRASTIC index, the acronym of 7 geological, hydrological and hydrogeological parameters, which collectively represents intrinsic (or natural) vulnerability and gives a sense of contaminants, such as nitrate-N, penetrating aquifers from the surface. These models are trained to modify or condition their DRASTIC index values by measured nitrate-N concentration. The three AI-techniques often perform similarly but have differences as well and therefore SICM exploits the situation to improve the modeled values by producing a hybrid modeling results through selecting better performing SVM, NF and GEP components. The models of the study area at Ardabil aquifer show that the vulnerability indices by the DRASTIC framework produces sharp fronts but AI models smoothen the fronts and reflect a better correlation with observed nitrate values; SICM improves on the performances of three AI models and cope well with heterogeneity and uncertain parameters. Copyright © 2016 Elsevier B.V. All rights reserved.

How Neural Networks Learn from Experience.

ERIC Educational Resources Information Center

Hinton, Geoffrey E.

1992-01-01

Discusses computational studies of learning in artificial neural networks and findings that may provide insights into the learning abilities of the human brain. Describes efforts to test theories about brain information processing, using artificial neural networks. Vignettes include information concerning how a neural network represents…

Neural network to diagnose lining condition

NASA Astrophysics Data System (ADS)

Yemelyanov, V. A.; Yemelyanova, N. Y.; Nedelkin, A. A.; Zarudnaya, M. V.

2018-03-01

The paper presents data on the problem of diagnosing the lining condition at the iron and steel works. The authors describe the neural network structure and software that are designed and developed to determine the lining burnout zones. The simulation results of the proposed neural networks are presented. The authors note the low learning and classification errors of the proposed neural networks. To realize the proposed neural network, the specialized software has been developed.

[Measurement and performance analysis of functional neural network].

PubMed

Li, Shan; Liu, Xinyu; Chen, Yan; Wan, Hong

2018-04-01

The measurement of network is one of the important researches in resolving neuronal population information processing mechanism using complex network theory. For the quantitative measurement problem of functional neural network, the relation between the measure indexes, i.e. the clustering coefficient, the global efficiency, the characteristic path length and the transitivity, and the network topology was analyzed. Then, the spike-based functional neural network was established and the simulation results showed that the measured network could represent the original neural connections among neurons. On the basis of the former work, the coding of functional neural network in nidopallium caudolaterale (NCL) about pigeon's motion behaviors was studied. We found that the NCL functional neural network effectively encoded the motion behaviors of the pigeon, and there were significant differences in four indexes among the left-turning, the forward and the right-turning. Overall, the establishment method of spike-based functional neural network is available and it is an effective tool to parse the brain information processing mechanism.

Neural network error correction for solving coupled ordinary differential equations

NASA Technical Reports Server (NTRS)

Shelton, R. O.; Darsey, J. A.; Sumpter, B. G.; Noid, D. W.

1992-01-01

A neural network is presented to learn errors generated by a numerical algorithm for solving coupled nonlinear differential equations. The method is based on using a neural network to correctly learn the error generated by, for example, Runge-Kutta on a model molecular dynamics (MD) problem. The neural network programs used in this study were developed by NASA. Comparisons are made for training the neural network using backpropagation and a new method which was found to converge with fewer iterations. The neural net programs, the MD model and the calculations are discussed.

Artificial and Bayesian Neural Networks

PubMed

Korhani Kangi, Azam; Bahrampour, Abbas

2018-02-26

Introduction and purpose: In recent years the use of neural networks without any premises for investigation of prognosis in analyzing survival data has increased. Artificial neural networks (ANN) use small processors with a continuous network to solve problems inspired by the human brain. Bayesian neural networks (BNN) constitute a neural-based approach to modeling and non-linearization of complex issues using special algorithms and statistical methods. Gastric cancer incidence is the first and third ranking for men and women in Iran, respectively. The aim of the present study was to assess the value of an artificial neural network and a Bayesian neural network for modeling and predicting of probability of gastric cancer patient death. Materials and Methods: In this study, we used information on 339 patients aged from 20 to 90 years old with positive gastric cancer, referred to Afzalipoor and Shahid Bahonar Hospitals in Kerman City from 2001 to 2015. The three layers perceptron neural network (ANN) and the Bayesian neural network (BNN) were used for predicting the probability of mortality using the available data. To investigate differences between the models, sensitivity, specificity, accuracy and the area under receiver operating characteristic curves (AUROCs) were generated. Results: In this study, the sensitivity and specificity of the artificial neural network and Bayesian neural network models were 0.882, 0.903 and 0.954, 0.909, respectively. Prediction accuracy and the area under curve ROC for the two models were 0.891, 0.944 and 0.935, 0.961. The age at diagnosis of gastric cancer was most important for predicting survival, followed by tumor grade, morphology, gender, smoking history, opium consumption, receiving chemotherapy, presence of metastasis, tumor stage, receiving radiotherapy, and being resident in a village. Conclusion: The findings of the present study indicated that the Bayesian neural network is preferable to an artificial neural network for predicting survival of gastric cancer patients in Iran. Creative Commons Attribution License

A comparative study of deep learning models for medical image classification

NASA Astrophysics Data System (ADS)

Dutta, Suvajit; Manideep, B. C. S.; Rai, Shalva; Vijayarajan, V.

2017-11-01

Deep Learning(DL) techniques are conquering over the prevailing traditional approaches of neural network, when it comes to the huge amount of dataset, applications requiring complex functions demanding increase accuracy with lower time complexities. Neurosciences has already exploited DL techniques, thus portrayed itself as an inspirational source for researchers exploring the domain of Machine learning. DL enthusiasts cover the areas of vision, speech recognition, motion planning and NLP as well, moving back and forth among fields. This concerns with building models that can successfully solve variety of tasks requiring intelligence and distributed representation. The accessibility to faster CPUs, introduction of GPUs-performing complex vector and matrix computations, supported agile connectivity to network. Enhanced software infrastructures for distributed computing worked in strengthening the thought that made researchers suffice DL methodologies. The paper emphases on the following DL procedures to traditional approaches which are performed manually for classifying medical images. The medical images are used for the study Diabetic Retinopathy(DR) and computed tomography (CT) emphysema data. Both DR and CT data diagnosis is difficult task for normal image classification methods. The initial work was carried out with basic image processing along with K-means clustering for identification of image severity levels. After determining image severity levels ANN has been applied on the data to get the basic classification result, then it is compared with the result of DNNs (Deep Neural Networks), which performed efficiently because of its multiple hidden layer features basically which increases accuracy factors, but the problem of vanishing gradient in DNNs made to consider Convolution Neural Networks (CNNs) as well for better results. The CNNs are found to be providing better outcomes when compared to other learning models aimed at classification of images. CNNs are favoured as they provide better visual processing models successfully classifying the noisy data as well. The work centres on the detection on Diabetic Retinopathy-loss in vision and recognition of computed tomography (CT) emphysema data measuring the severity levels for both cases. The paper discovers how various Machine Learning algorithms can be implemented ensuing a supervised approach, so as to get accurate results with less complexity possible.

Model Of Neural Network With Creative Dynamics

NASA Technical Reports Server (NTRS)

Zak, Michail; Barhen, Jacob

1993-01-01

Paper presents analysis of mathematical model of one-neuron/one-synapse neural network featuring coupled activation and learning dynamics and parametrical periodic excitation. Demonstrates self-programming, partly random behavior of suitable designed neural network; believed to be related to spontaneity and creativity of biological neural networks.

Low-dimensional recurrent neural network-based Kalman filter for speech enhancement.

PubMed

Xia, Youshen; Wang, Jun

2015-07-01

This paper proposes a new recurrent neural network-based Kalman filter for speech enhancement, based on a noise-constrained least squares estimate. The parameters of speech signal modeled as autoregressive process are first estimated by using the proposed recurrent neural network and the speech signal is then recovered from Kalman filtering. The proposed recurrent neural network is globally asymptomatically stable to the noise-constrained estimate. Because the noise-constrained estimate has a robust performance against non-Gaussian noise, the proposed recurrent neural network-based speech enhancement algorithm can minimize the estimation error of Kalman filter parameters in non-Gaussian noise. Furthermore, having a low-dimensional model feature, the proposed neural network-based speech enhancement algorithm has a much faster speed than two existing recurrent neural networks-based speech enhancement algorithms. Simulation results show that the proposed recurrent neural network-based speech enhancement algorithm can produce a good performance with fast computation and noise reduction. Copyright © 2015 Elsevier Ltd. All rights reserved.

Voxel-Based Neighborhood for Spatial Shape Pattern Classification of Lidar Point Clouds with Supervised Learning

PubMed Central

Plaza-Leiva, Victoria; Gomez-Ruiz, Jose Antonio; Mandow, Anthony; García-Cerezo, Alfonso

2017-01-01

Improving the effectiveness of spatial shape features classification from 3D lidar data is very relevant because it is largely used as a fundamental step towards higher level scene understanding challenges of autonomous vehicles and terrestrial robots. In this sense, computing neighborhood for points in dense scans becomes a costly process for both training and classification. This paper proposes a new general framework for implementing and comparing different supervised learning classifiers with a simple voxel-based neighborhood computation where points in each non-overlapping voxel in a regular grid are assigned to the same class by considering features within a support region defined by the voxel itself. The contribution provides offline training and online classification procedures as well as five alternative feature vector definitions based on principal component analysis for scatter, tubular and planar shapes. Moreover, the feasibility of this approach is evaluated by implementing a neural network (NN) method previously proposed by the authors as well as three other supervised learning classifiers found in scene processing methods: support vector machines (SVM), Gaussian processes (GP), and Gaussian mixture models (GMM). A comparative performance analysis is presented using real point clouds from both natural and urban environments and two different 3D rangefinders (a tilting Hokuyo UTM-30LX and a Riegl). Classification performance metrics and processing time measurements confirm the benefits of the NN classifier and the feasibility of voxel-based neighborhood. PMID:28294963

Deep Learning Representation from Electroencephalography of Early-Stage Creutzfeldt-Jakob Disease and Features for Differentiation from Rapidly Progressive Dementia.

PubMed

Morabito, Francesco Carlo; Campolo, Maurizio; Mammone, Nadia; Versaci, Mario; Franceschetti, Silvana; Tagliavini, Fabrizio; Sofia, Vito; Fatuzzo, Daniela; Gambardella, Antonio; Labate, Angelo; Mumoli, Laura; Tripodi, Giovanbattista Gaspare; Gasparini, Sara; Cianci, Vittoria; Sueri, Chiara; Ferlazzo, Edoardo; Aguglia, Umberto

2017-03-01

A novel technique of quantitative EEG for differentiating patients with early-stage Creutzfeldt-Jakob disease (CJD) from other forms of rapidly progressive dementia (RPD) is proposed. The discrimination is based on the extraction of suitable features from the time-frequency representation of the EEG signals through continuous wavelet transform (CWT). An average measure of complexity of the EEG signal obtained by permutation entropy (PE) is also included. The dimensionality of the feature space is reduced through a multilayer processing system based on the recently emerged deep learning (DL) concept. The DL processor includes a stacked auto-encoder, trained by unsupervised learning techniques, and a classifier whose parameters are determined in a supervised way by associating the known category labels to the reduced vector of high-level features generated by the previous processing blocks. The supervised learning step is carried out by using either support vector machines (SVM) or multilayer neural networks (MLP-NN). A subset of EEG from patients suffering from Alzheimer's Disease (AD) and healthy controls (HC) is considered for differentiating CJD patients. When fine-tuning the parameters of the global processing system by a supervised learning procedure, the proposed system is able to achieve an average accuracy of 89%, an average sensitivity of 92%, and an average specificity of 89% in differentiating CJD from RPD. Similar results are obtained for CJD versus AD and CJD versus HC.

Creating a Reinforcement Learning Controller for Functional Electrical Stimulation of a Human Arm*

PubMed Central

Thomas, Philip S.; Branicky, Michael; van den Bogert, Antonie; Jagodnik, Kathleen

2010-01-01

Clinical tests have shown that the dynamics of a human arm, controlled using Functional Electrical Stimulation (FES), can vary significantly between and during trials. In this paper, we study the application of Reinforcement Learning to create a controller that can adapt to these changing dynamics of a human arm. Development and tests were done in simulation using a two-dimensional arm model and Hill-based muscle dynamics. An actor-critic architecture is used with artificial neural networks for both the actor and the critic. We begin by training it using a Proportional Derivative (PD) controller as a supervisor. We then make clinically relevant changes to the dynamics of the arm and test the actor-critic’s ability to adapt without supervision in a reasonable number of episodes. PMID:22081795

Thermalnet: a Deep Convolutional Network for Synthetic Thermal Image Generation

NASA Astrophysics Data System (ADS)

Kniaz, V. V.; Gorbatsevich, V. S.; Mizginov, V. A.

2017-05-01

Deep convolutional neural networks have dramatically changed the landscape of the modern computer vision. Nowadays methods based on deep neural networks show the best performance among image recognition and object detection algorithms. While polishing of network architectures received a lot of scholar attention, from the practical point of view the preparation of a large image dataset for a successful training of a neural network became one of major challenges. This challenge is particularly profound for image recognition in wavelengths lying outside the visible spectrum. For example no infrared or radar image datasets large enough for successful training of a deep neural network are available to date in public domain. Recent advances of deep neural networks prove that they are also capable to do arbitrary image transformations such as super-resolution image generation, grayscale image colorisation and imitation of style of a given artist. Thus a natural question arise: how could be deep neural networks used for augmentation of existing large image datasets? This paper is focused on the development of the Thermalnet deep convolutional neural network for augmentation of existing large visible image datasets with synthetic thermal images. The Thermalnet network architecture is inspired by colorisation deep neural networks.

Determining geophysical properties from well log data using artificial neural networks and fuzzy inference systems

NASA Astrophysics Data System (ADS)

Chang, Hsien-Cheng

Two novel synergistic systems consisting of artificial neural networks and fuzzy inference systems are developed to determine geophysical properties by using well log data. These systems are employed to improve the determination accuracy in carbonate rocks, which are generally more complex than siliciclastic rocks. One system, consisting of a single adaptive resonance theory (ART) neural network and three fuzzy inference systems (FISs), is used to determine the permeability category. The other system, which is composed of three ART neural networks and a single FIS, is employed to determine the lithofacies. The geophysical properties studied in this research, permeability category and lithofacies, are treated as categorical data. The permeability values are transformed into a "permeability category" to account for the effects of scale differences between core analyses and well logs, and heterogeneity in the carbonate rocks. The ART neural networks dynamically cluster the input data sets into different groups. The FIS is used to incorporate geologic experts' knowledge, which is usually in linguistic forms, into systems. These synergistic systems thus provide viable alternative solutions to overcome the effects of heterogeneity, the uncertainties of carbonate rock depositional environments, and the scarcity of well log data. The results obtained in this research show promising improvements over backpropagation neural networks. For the permeability category, the prediction accuracies are 68.4% and 62.8% for the multiple-single ART neural network-FIS and a single backpropagation neural network, respectively. For lithofacies, the prediction accuracies are 87.6%, 79%, and 62.8% for the single-multiple ART neural network-FIS, a single ART neural network, and a single backpropagation neural network, respectively. The sensitivity analysis results show that the multiple-single ART neural networks-FIS and a single ART neural network possess the same matching trends in determining lithofacies. This research shows that the adaptive resonance theory neural networks enable decision-makers to clearly distinguish the importance of different pieces of data which are useful in three-dimensional subsurface modeling. Geologic experts' knowledge can be easily applied and maintained by using the fuzzy inference systems.

Reducing neural network training time with parallel processing

NASA Technical Reports Server (NTRS)

Rogers, James L., Jr.; Lamarsh, William J., II

1995-01-01

Obtaining optimal solutions for engineering design problems is often expensive because the process typically requires numerous iterations involving analysis and optimization programs. Previous research has shown that a near optimum solution can be obtained in less time by simulating a slow, expensive analysis with a fast, inexpensive neural network. A new approach has been developed to further reduce this time. This approach decomposes a large neural network into many smaller neural networks that can be trained in parallel. Guidelines are developed to avoid some of the pitfalls when training smaller neural networks in parallel. These guidelines allow the engineer: to determine the number of nodes on the hidden layer of the smaller neural networks; to choose the initial training weights; and to select a network configuration that will capture the interactions among the smaller neural networks. This paper presents results describing how these guidelines are developed.

Application of the ANNA neural network chip to high-speed character recognition.

PubMed

Sackinger, E; Boser, B E; Bromley, J; Lecun, Y; Jackel, L D

1992-01-01

A neural network with 136000 connections for recognition of handwritten digits has been implemented using a mixed analog/digital neural network chip. The neural network chip is capable of processing 1000 characters/s. The recognition system has essentially the same rate (5%) as a simulation of the network with 32-b floating-point precision.

Integration of Gravitational Torques in Cerebellar Pathways Allows for the Dynamic Inverse Computation of Vertical Pointing Movements of a Robot Arm

PubMed Central

Gentili, Rodolphe J.; Papaxanthis, Charalambos; Ebadzadeh, Mehdi; Eskiizmirliler, Selim; Ouanezar, Sofiane; Darlot, Christian

2009-01-01

Background Several authors suggested that gravitational forces are centrally represented in the brain for planning, control and sensorimotor predictions of movements. Furthermore, some studies proposed that the cerebellum computes the inverse dynamics (internal inverse model) whereas others suggested that it computes sensorimotor predictions (internal forward model). Methodology/Principal Findings This study proposes a model of cerebellar pathways deduced from both biological and physical constraints. The model learns the dynamic inverse computation of the effect of gravitational torques from its sensorimotor predictions without calculating an explicit inverse computation. By using supervised learning, this model learns to control an anthropomorphic robot arm actuated by two antagonists McKibben artificial muscles. This was achieved by using internal parallel feedback loops containing neural networks which anticipate the sensorimotor consequences of the neural commands. The artificial neural networks architecture was similar to the large-scale connectivity of the cerebellar cortex. Movements in the sagittal plane were performed during three sessions combining different initial positions, amplitudes and directions of movements to vary the effects of the gravitational torques applied to the robotic arm. The results show that this model acquired an internal representation of the gravitational effects during vertical arm pointing movements. Conclusions/Significance This is consistent with the proposal that the cerebellar cortex contains an internal representation of gravitational torques which is encoded through a learning process. Furthermore, this model suggests that the cerebellum performs the inverse dynamics computation based on sensorimotor predictions. This highlights the importance of sensorimotor predictions of gravitational torques acting on upper limb movements performed in the gravitational field. PMID:19384420

Machine Learning and Quantum Mechanics

NASA Astrophysics Data System (ADS)

Chapline, George

The author has previously pointed out some similarities between selforganizing neural networks and quantum mechanics. These types of neural networks were originally conceived of as away of emulating the cognitive capabilities of the human brain. Recently extensions of these networks, collectively referred to as deep learning networks, have strengthened the connection between self-organizing neural networks and human cognitive capabilities. In this note we consider whether hardware quantum devices might be useful for emulating neural networks with human-like cognitive capabilities, or alternatively whether implementations of deep learning neural networks using conventional computers might lead to better algorithms for solving the many body Schrodinger equation.

Networks of Professional Supervision

ERIC Educational Resources Information Center

Annan, Jean; Ryba, Ken

2013-01-01

An ecological analysis of the supervisory activity of 31 New Zealand school psychologists examined simultaneously the theories of school psychology, supervision practices, and the contextual qualities that mediated participants' supervisory actions. The findings indicated that the school psychologists worked to achieve the supervision goals of…

Using fuzzy logic to integrate neural networks and knowledge-based systems

NASA Technical Reports Server (NTRS)

Yen, John

1991-01-01

Outlined here is a novel hybrid architecture that uses fuzzy logic to integrate neural networks and knowledge-based systems. The author's approach offers important synergistic benefits to neural nets, approximate reasoning, and symbolic processing. Fuzzy inference rules extend symbolic systems with approximate reasoning capabilities, which are used for integrating and interpreting the outputs of neural networks. The symbolic system captures meta-level information about neural networks and defines its interaction with neural networks through a set of control tasks. Fuzzy action rules provide a robust mechanism for recognizing the situations in which neural networks require certain control actions. The neural nets, on the other hand, offer flexible classification and adaptive learning capabilities, which are crucial for dynamic and noisy environments. By combining neural nets and symbolic systems at their system levels through the use of fuzzy logic, the author's approach alleviates current difficulties in reconciling differences between low-level data processing mechanisms of neural nets and artificial intelligence systems.

A neural network application to classification of health status of HIV/AIDS patients.

PubMed

Kwak, N K; Lee, C

1997-04-01

This paper presents an application of neural networks to classify and to predict the health status of HIV/AIDS patients. A neural network model in classifying both the well and not-well health status of HIV/AIDS patients is developed and evaluated in terms of validity and reliability of the test. Several different neural network topologies are applied to AIDS Cost and Utilization Survey (ACSUS) datasets in order to demonstrate the neural network's capability.

Deep Learning Neural Networks and Bayesian Neural Networks in Data Analysis

NASA Astrophysics Data System (ADS)

Chernoded, Andrey; Dudko, Lev; Myagkov, Igor; Volkov, Petr

2017-10-01

Most of the modern analyses in high energy physics use signal-versus-background classification techniques of machine learning methods and neural networks in particular. Deep learning neural network is the most promising modern technique to separate signal and background and now days can be widely and successfully implemented as a part of physical analysis. In this article we compare Deep learning and Bayesian neural networks application as a classifiers in an instance of top quark analysis.

Deep Supervised, but Not Unsupervised, Models May Explain IT Cortical Representation

PubMed Central

Khaligh-Razavi, Seyed-Mahdi; Kriegeskorte, Nikolaus

2014-01-01

Inferior temporal (IT) cortex in human and nonhuman primates serves visual object recognition. Computational object-vision models, although continually improving, do not yet reach human performance. It is unclear to what extent the internal representations of computational models can explain the IT representation. Here we investigate a wide range of computational model representations (37 in total), testing their categorization performance and their ability to account for the IT representational geometry. The models include well-known neuroscientific object-recognition models (e.g. HMAX, VisNet) along with several models from computer vision (e.g. SIFT, GIST, self-similarity features, and a deep convolutional neural network). We compared the representational dissimilarity matrices (RDMs) of the model representations with the RDMs obtained from human IT (measured with fMRI) and monkey IT (measured with cell recording) for the same set of stimuli (not used in training the models). Better performing models were more similar to IT in that they showed greater clustering of representational patterns by category. In addition, better performing models also more strongly resembled IT in terms of their within-category representational dissimilarities. Representational geometries were significantly correlated between IT and many of the models. However, the categorical clustering observed in IT was largely unexplained by the unsupervised models. The deep convolutional network, which was trained by supervision with over a million category-labeled images, reached the highest categorization performance and also best explained IT, although it did not fully explain the IT data. Combining the features of this model with appropriate weights and adding linear combinations that maximize the margin between animate and inanimate objects and between faces and other objects yielded a representation that fully explained our IT data. Overall, our results suggest that explaining IT requires computational features trained through supervised learning to emphasize the behaviorally important categorical divisions prominently reflected in IT. PMID:25375136

Improvement of the Hopfield Neural Network by MC-Adaptation Rule

NASA Astrophysics Data System (ADS)

Zhou, Zhen; Zhao, Hong

2006-06-01

We show that the performance of the Hopfield neural networks, especially the quality of the recall and the capacity of the effective storing, can be greatly improved by making use of a recently presented neural network designing method without altering the whole structure of the network. In the improved neural network, a memory pattern is recalled exactly from initial states having a given degree of similarity with the memory pattern, and thus one can avoids to apply the overlap criterion as carried out in the Hopfield neural networks.

The Energy Coding of a Structural Neural Network Based on the Hodgkin-Huxley Model.

PubMed

Zhu, Zhenyu; Wang, Rubin; Zhu, Fengyun

2018-01-01

Based on the Hodgkin-Huxley model, the present study established a fully connected structural neural network to simulate the neural activity and energy consumption of the network by neural energy coding theory. The numerical simulation result showed that the periodicity of the network energy distribution was positively correlated to the number of neurons and coupling strength, but negatively correlated to signal transmitting delay. Moreover, a relationship was established between the energy distribution feature and the synchronous oscillation of the neural network, which showed that when the proportion of negative energy in power consumption curve was high, the synchronous oscillation of the neural network was apparent. In addition, comparison with the simulation result of structural neural network based on the Wang-Zhang biophysical model of neurons showed that both models were essentially consistent.

An adaptive Hinfinity controller design for bank-to-turn missiles using ridge Gaussian neural networks.

PubMed

Lin, Chuan-Kai; Wang, Sheng-De

2004-11-01

A new autopilot design for bank-to-turn (BTT) missiles is presented. In the design of autopilot, a ridge Gaussian neural network with local learning capability and fewer tuning parameters than Gaussian neural networks is proposed to model the controlled nonlinear systems. We prove that the proposed ridge Gaussian neural network, which can be a universal approximator, equals the expansions of rotated and scaled Gaussian functions. Although ridge Gaussian neural networks can approximate the nonlinear and complex systems accurately, the small approximation errors may affect the tracking performance significantly. Therefore, by employing the Hinfinity control theory, it is easy to attenuate the effects of the approximation errors of the ridge Gaussian neural networks to a prescribed level. Computer simulation results confirm the effectiveness of the proposed ridge Gaussian neural networks-based autopilot with Hinfinity stabilization.

Constraint satisfaction adaptive neural network and heuristics combined approaches for generalized job-shop scheduling.

PubMed

Yang, S; Wang, D

2000-01-01

This paper presents a constraint satisfaction adaptive neural network, together with several heuristics, to solve the generalized job-shop scheduling problem, one of NP-complete constraint satisfaction problems. The proposed neural network can be easily constructed and can adaptively adjust its weights of connections and biases of units based on the sequence and resource constraints of the job-shop scheduling problem during its processing. Several heuristics that can be combined with the neural network are also presented. In the combined approaches, the neural network is used to obtain feasible solutions, the heuristic algorithms are used to improve the performance of the neural network and the quality of the obtained solutions. Simulations have shown that the proposed neural network and its combined approaches are efficient with respect to the quality of solutions and the solving speed.

Financial time series prediction using spiking neural networks.

PubMed

Reid, David; Hussain, Abir Jaafar; Tawfik, Hissam

2014-01-01

In this paper a novel application of a particular type of spiking neural network, a Polychronous Spiking Network, was used for financial time series prediction. It is argued that the inherent temporal capabilities of this type of network are suited to non-stationary data such as this. The performance of the spiking neural network was benchmarked against three systems: two "traditional", rate-encoded, neural networks; a Multi-Layer Perceptron neural network and a Dynamic Ridge Polynomial neural network, and a standard Linear Predictor Coefficients model. For this comparison three non-stationary and noisy time series were used: IBM stock data; US/Euro exchange rate data, and the price of Brent crude oil. The experiments demonstrated favourable prediction results for the Spiking Neural Network in terms of Annualised Return and prediction error for 5-Step ahead predictions. These results were also supported by other relevant metrics such as Maximum Drawdown and Signal-To-Noise ratio. This work demonstrated the applicability of the Polychronous Spiking Network to financial data forecasting and this in turn indicates the potential of using such networks over traditional systems in difficult to manage non-stationary environments.

Non-Intrusive Gaze Tracking Using Artificial Neural Networks

DTIC Science & Technology

1994-01-05

We have developed an artificial neural network based gaze tracking, system which can be customized to individual users. A three layer feed forward...empirical analysis of the performance of a large number of artificial neural network architectures for this task. Suggestions for further explorations...for neurally based gaze trackers are presented, and are related to other similar artificial neural network applications such as autonomous road following.

Neural dynamics based on the recognition of neural fingerprints

PubMed Central

Carrillo-Medina, José Luis; Latorre, Roberto

2015-01-01

Experimental evidence has revealed the existence of characteristic spiking features in different neural signals, e.g., individual neural signatures identifying the emitter or functional signatures characterizing specific tasks. These neural fingerprints may play a critical role in neural information processing, since they allow receptors to discriminate or contextualize incoming stimuli. This could be a powerful strategy for neural systems that greatly enhances the encoding and processing capacity of these networks. Nevertheless, the study of information processing based on the identification of specific neural fingerprints has attracted little attention. In this work, we study (i) the emerging collective dynamics of a network of neurons that communicate with each other by exchange of neural fingerprints and (ii) the influence of the network topology on the self-organizing properties within the network. Complex collective dynamics emerge in the network in the presence of stimuli. Predefined inputs, i.e., specific neural fingerprints, are detected and encoded into coexisting patterns of activity that propagate throughout the network with different spatial organization. The patterns evoked by a stimulus can survive after the stimulation is over, which provides memory mechanisms to the network. The results presented in this paper suggest that neural information processing based on neural fingerprints can be a plausible, flexible, and powerful strategy. PMID:25852531

Structural reliability calculation method based on the dual neural network and direct integration method.

PubMed

Li, Haibin; He, Yun; Nie, Xiaobo

2018-01-01

Structural reliability analysis under uncertainty is paid wide attention by engineers and scholars due to reflecting the structural characteristics and the bearing actual situation. The direct integration method, started from the definition of reliability theory, is easy to be understood, but there are still mathematics difficulties in the calculation of multiple integrals. Therefore, a dual neural network method is proposed for calculating multiple integrals in this paper. Dual neural network consists of two neural networks. The neural network A is used to learn the integrand function, and the neural network B is used to simulate the original function. According to the derivative relationships between the network output and the network input, the neural network B is derived from the neural network A. On this basis, the performance function of normalization is employed in the proposed method to overcome the difficulty of multiple integrations and to improve the accuracy for reliability calculations. The comparisons between the proposed method and Monte Carlo simulation method, Hasofer-Lind method, the mean value first-order second moment method have demonstrated that the proposed method is an efficient and accurate reliability method for structural reliability problems.

Patterns of synchrony for feed-forward and auto-regulation feed-forward neural networks.

PubMed

Aguiar, Manuela A D; Dias, Ana Paula S; Ferreira, Flora

2017-01-01

We consider feed-forward and auto-regulation feed-forward neural (weighted) coupled cell networks. In feed-forward neural networks, cells are arranged in layers such that the cells of the first layer have empty input set and cells of each other layer receive only inputs from cells of the previous layer. An auto-regulation feed-forward neural coupled cell network is a feed-forward neural network where additionally some cells of the first layer have auto-regulation, that is, they have a self-loop. Given a network structure, a robust pattern of synchrony is a space defined in terms of equalities of cell coordinates that is flow-invariant for any coupled cell system (with additive input structure) associated with the network. In this paper, we describe the robust patterns of synchrony for feed-forward and auto-regulation feed-forward neural networks. Regarding feed-forward neural networks, we show that only cells in the same layer can synchronize. On the other hand, in the presence of auto-regulation, we prove that cells in different layers can synchronize in a robust way and we give a characterization of the possible patterns of synchrony that can occur for auto-regulation feed-forward neural networks.

Pattern classification and recognition of invertebrate functional groups using self-organizing neural networks.

PubMed

Zhang, WenJun

2007-07-01

Self-organizing neural networks can be used to mimic non-linear systems. The main objective of this study is to make pattern classification and recognition on sampling information using two self-organizing neural network models. Invertebrate functional groups sampled in the irrigated rice field were classified and recognized using one-dimensional self-organizing map and self-organizing competitive learning neural networks. Comparisons between neural network models, distance (similarity) measures, and number of neurons were conducted. The results showed that self-organizing map and self-organizing competitive learning neural network models were effective in pattern classification and recognition of sampling information. Overall the performance of one-dimensional self-organizing map neural network was better than self-organizing competitive learning neural network. The number of neurons could determine the number of classes in the classification. Different neural network models with various distance (similarity) measures yielded similar classifications. Some differences, dependent upon the specific network structure, would be found. The pattern of an unrecognized functional group was recognized with the self-organizing neural network. A relative consistent classification indicated that the following invertebrate functional groups, terrestrial blood sucker; terrestrial flyer; tourist (nonpredatory species with no known functional role other than as prey in ecosystem); gall former; collector (gather, deposit feeder); predator and parasitoid; leaf miner; idiobiont (acarine ectoparasitoid), were classified into the same group, and the following invertebrate functional groups, external plant feeder; terrestrial crawler, walker, jumper or hunter; neustonic (water surface) swimmer (semi-aquatic), were classified into another group. It was concluded that reliable conclusions could be drawn from comparisons of different neural network models that use different distance (similarity) measures. Results with the larger consistency will be more reliable.

Thermoelastic steam turbine rotor control based on neural network

NASA Astrophysics Data System (ADS)

Rzadkowski, Romuald; Dominiczak, Krzysztof; Radulski, Wojciech; Szczepanik, R.

2015-12-01

Considered here are Nonlinear Auto-Regressive neural networks with eXogenous inputs (NARX) as a mathematical model of a steam turbine rotor for controlling steam turbine stress on-line. In order to obtain neural networks that locate critical stress and temperature points in the steam turbine during transient states, an FE rotor model was built. This model was used to train the neural networks on the basis of steam turbine transient operating data. The training included nonlinearity related to steam turbine expansion, heat exchange and rotor material properties during transients. Simultaneous neural networks are algorithms which can be implemented on PLC controllers. This allows for the application neural networks to control steam turbine stress in industrial power plants.

The use of artificial neural networks in experimental data acquisition and aerodynamic design

NASA Technical Reports Server (NTRS)

Meade, Andrew J., Jr.

1991-01-01

It is proposed that an artificial neural network be used to construct an intelligent data acquisition system. The artificial neural networks (ANN) model has a potential for replacing traditional procedures as well as for use in computational fluid dynamics validation. Potential advantages of the ANN model are listed. As a proof of concept, the author modeled a NACA 0012 airfoil at specific conditions, using the neural network simulator NETS, developed by James Baffes of the NASA Johnson Space Center. The neural network predictions were compared to the actual data. It is concluded that artificial neural networks can provide an elegant and valuable class of mathematical tools for data analysis.

Research on artificial neural network intrusion detection photochemistry based on the improved wavelet analysis and transformation

NASA Astrophysics Data System (ADS)

Li, Hong; Ding, Xue

2017-03-01

This paper combines wavelet analysis and wavelet transform theory with artificial neural network, through the pretreatment on point feature attributes before in intrusion detection, to make them suitable for improvement of wavelet neural network. The whole intrusion classification model gets the better adaptability, self-learning ability, greatly enhances the wavelet neural network for solving the problem of field detection invasion, reduces storage space, contributes to improve the performance of the constructed neural network, and reduces the training time. Finally the results of the KDDCup99 data set simulation experiment shows that, this method reduces the complexity of constructing wavelet neural network, but also ensures the accuracy of the intrusion classification.

A class of finite-time dual neural networks for solving quadratic programming problems and its k-winners-take-all application.

PubMed

Li, Shuai; Li, Yangming; Wang, Zheng

2013-03-01

This paper presents a class of recurrent neural networks to solve quadratic programming problems. Different from most existing recurrent neural networks for solving quadratic programming problems, the proposed neural network model converges in finite time and the activation function is not required to be a hard-limiting function for finite convergence time. The stability, finite-time convergence property and the optimality of the proposed neural network for solving the original quadratic programming problem are proven in theory. Extensive simulations are performed to evaluate the performance of the neural network with different parameters. In addition, the proposed neural network is applied to solving the k-winner-take-all (k-WTA) problem. Both theoretical analysis and numerical simulations validate the effectiveness of our method for solving the k-WTA problem. Copyright © 2012 Elsevier Ltd. All rights reserved.

Satellite image analysis using neural networks

NASA Technical Reports Server (NTRS)

Sheldon, Roger A.

1990-01-01

The tremendous backlog of unanalyzed satellite data necessitates the development of improved methods for data cataloging and analysis. Ford Aerospace has developed an image analysis system, SIANN (Satellite Image Analysis using Neural Networks) that integrates the technologies necessary to satisfy NASA's science data analysis requirements for the next generation of satellites. SIANN will enable scientists to train a neural network to recognize image data containing scenes of interest and then rapidly search data archives for all such images. The approach combines conventional image processing technology with recent advances in neural networks to provide improved classification capabilities. SIANN allows users to proceed through a four step process of image classification: filtering and enhancement, creation of neural network training data via application of feature extraction algorithms, configuring and training a neural network model, and classification of images by application of the trained neural network. A prototype experimentation testbed was completed and applied to climatological data.

Firing patterns transition and desynchronization induced by time delay in neural networks

NASA Astrophysics Data System (ADS)

Huang, Shoufang; Zhang, Jiqian; Wang, Maosheng; Hu, Chin-Kun

2018-06-01

We used the Hindmarsh-Rose (HR) model (Hindmarsh and Rose, 1984) to study the effect of time delay on the transition of firing behaviors and desynchronization in neural networks. As time delay is increased, neural networks exhibit diversity of firing behaviors, including regular spiking or bursting and firing patterns transitions (FPTs). Meanwhile, the desynchronization of firing and unstable bursting with decreasing amplitude in neural system, are also increasingly enhanced with the increase of time delay. Furthermore, we also studied the effect of coupling strength and network randomness on these phenomena. Our results imply that time delays can induce transition and desynchronization of firing behaviors in neural networks. These findings provide new insight into the role of time delay in the firing activities of neural networks, and can help to better understand the firing phenomena in complex systems of neural networks. A possible mechanism in brain that can cause the increase of time delay is discussed.

A one-layer recurrent neural network for constrained pseudoconvex optimization and its application for dynamic portfolio optimization.

PubMed

Liu, Qingshan; Guo, Zhishan; Wang, Jun

2012-02-01

In this paper, a one-layer recurrent neural network is proposed for solving pseudoconvex optimization problems subject to linear equality and bound constraints. Compared with the existing neural networks for optimization (e.g., the projection neural networks), the proposed neural network is capable of solving more general pseudoconvex optimization problems with equality and bound constraints. Moreover, it is capable of solving constrained fractional programming problems as a special case. The convergence of the state variables of the proposed neural network to achieve solution optimality is guaranteed as long as the designed parameters in the model are larger than the derived lower bounds. Numerical examples with simulation results illustrate the effectiveness and characteristics of the proposed neural network. In addition, an application for dynamic portfolio optimization is discussed. Copyright © 2011 Elsevier Ltd. All rights reserved.

Applications of artificial neural nets in clinical biomechanics.

PubMed

Schöllhorn, W I

2004-11-01

The purpose of this article is to provide an overview of current applications of artificial neural networks in the area of clinical biomechanics. The body of literature on artificial neural networks grew intractably vast during the last 15 years. Conventional statistical models may present certain limitations that can be overcome by neural networks. Artificial neural networks in general are introduced, some limitations, and some proven benefits are discussed.

Neural Networks for Rapid Design and Analysis

NASA Technical Reports Server (NTRS)

Sparks, Dean W., Jr.; Maghami, Peiman G.

1998-01-01

Artificial neural networks have been employed for rapid and efficient dynamics and control analysis of flexible systems. Specifically, feedforward neural networks are designed to approximate nonlinear dynamic components over prescribed input ranges, and are used in simulations as a means to speed up the overall time response analysis process. To capture the recursive nature of dynamic components with artificial neural networks, recurrent networks, which use state feedback with the appropriate number of time delays, as inputs to the networks, are employed. Once properly trained, neural networks can give very good approximations to nonlinear dynamic components, and by their judicious use in simulations, allow the analyst the potential to speed up the analysis process considerably. To illustrate this potential speed up, an existing simulation model of a spacecraft reaction wheel system is executed, first conventionally, and then with an artificial neural network in place.

Optimal input sizes for neural network de-interlacing

NASA Astrophysics Data System (ADS)

Choi, Hyunsoo; Seo, Guiwon; Lee, Chulhee

2009-02-01

Neural network de-interlacing has shown promising results among various de-interlacing methods. In this paper, we investigate the effects of input size for neural networks for various video formats when the neural networks are used for de-interlacing. In particular, we investigate optimal input sizes for CIF, VGA and HD video formats.

Impact of leakage delay on bifurcation in high-order fractional BAM neural networks.

PubMed

Huang, Chengdai; Cao, Jinde

2018-02-01

The effects of leakage delay on the dynamics of neural networks with integer-order have lately been received considerable attention. It has been confirmed that fractional neural networks more appropriately uncover the dynamical properties of neural networks, but the results of fractional neural networks with leakage delay are relatively few. This paper primarily concentrates on the issue of bifurcation for high-order fractional bidirectional associative memory(BAM) neural networks involving leakage delay. The first attempt is made to tackle the stability and bifurcation of high-order fractional BAM neural networks with time delay in leakage terms in this paper. The conditions for the appearance of bifurcation for the proposed systems with leakage delay are firstly established by adopting time delay as a bifurcation parameter. Then, the bifurcation criteria of such system without leakage delay are successfully acquired. Comparative analysis wondrously detects that the stability performance of the proposed high-order fractional neural networks is critically weakened by leakage delay, they cannot be overlooked. Numerical examples are ultimately exhibited to attest the efficiency of the theoretical results. Copyright © 2017 Elsevier Ltd. All rights reserved.

Coronary Artery Diagnosis Aided by Neural Network

NASA Astrophysics Data System (ADS)

Stefko, Kamil

2007-01-01

Coronary artery disease is due to atheromatous narrowing and subsequent occlusion of the coronary vessel. Application of optimised feed forward multi-layer back propagation neural network (MLBP) for detection of narrowing in coronary artery vessels is presented in this paper. The research was performed using 580 data records from traditional ECG exercise test confirmed by coronary arteriography results. Each record of training database included description of the state of a patient providing input data for the neural network. Level and slope of ST segment of a 12 lead ECG signal recorded at rest and after effort (48 floating point values) was the main component of input data for neural network was. Coronary arteriography results (verified the existence or absence of more than 50% stenosis of the particular coronary vessels) were used as a correct neural network training output pattern. More than 96% of cases were correctly recognised by especially optimised and a thoroughly verified neural network. Leave one out method was used for neural network verification so 580 data records could be used for training as well as for verification of neural network.

Predicate calculus for an architecture of multiple neural networks

NASA Astrophysics Data System (ADS)

Consoli, Robert H.

1990-08-01

Future projects with neural networks will require multiple individual network components. Current efforts along these lines are ad hoc. This paper relates the neural network to a classical device and derives a multi-part architecture from that model. Further it provides a Predicate Calculus variant for describing the location and nature of the trainings and suggests Resolution Refutation as a method for determining the performance of the system as well as the location of needed trainings for specific proofs. 2. THE NEURAL NETWORK AND A CLASSICAL DEVICE Recently investigators have been making reports about architectures of multiple neural networksL234. These efforts are appearing at an early stage in neural network investigations they are characterized by architectures suggested directly by the problem space. Touretzky and Hinton suggest an architecture for processing logical statements1 the design of this architecture arises from the syntax of a restricted class of logical expressions and exhibits syntactic limitations. In similar fashion a multiple neural netword arises out of a control problem2 from the sequence learning problem3 and from the domain of machine learning. 4 But a general theory of multiple neural devices is missing. More general attempts to relate single or multiple neural networks to classical computing devices are not common although an attempt is made to relate single neural devices to a Turing machines and Sun et a!. develop a multiple neural architecture that performs pattern classification.

Learning Data Set Influence on Identification Accuracy of Gas Turbine Neural Network Model

NASA Astrophysics Data System (ADS)

Kuznetsov, A. V.; Makaryants, G. M.

2018-01-01

There are many gas turbine engine identification researches via dynamic neural network models. It should minimize errors between model and real object during identification process. Questions about training data set processing of neural networks are usually missed. This article presents a study about influence of data set type on gas turbine neural network model accuracy. The identification object is thermodynamic model of micro gas turbine engine. The thermodynamic model input signal is the fuel consumption and output signal is the engine rotor rotation frequency. Four types input signals was used for creating training and testing data sets of dynamic neural network models - step, fast, slow and mixed. Four dynamic neural networks were created based on these types of training data sets. Each neural network was tested via four types test data sets. In the result 16 transition processes from four neural networks and four test data sets from analogous solving results of thermodynamic model were compared. The errors comparison was made between all neural network errors in each test data set. In the comparison result it was shown error value ranges of each test data set. It is shown that error values ranges is small therefore the influence of data set types on identification accuracy is low.

Altered Synchronizations among Neural Networks in Geriatric Depression

PubMed Central

Wang, Lihong; Chou, Ying-Hui; Potter, Guy G.; Steffens, David C.

2015-01-01

Although major depression has been considered as a manifestation of discoordinated activity between affective and cognitive neural networks, only a few studies have examined the relationships among neural networks directly. Because of the known disconnection theory, geriatric depression could be a useful model in studying the interactions among different networks. In the present study, using independent component analysis to identify intrinsically connected neural networks, we investigated the alterations in synchronizations among neural networks in geriatric depression to better understand the underlying neural mechanisms. Resting-state fMRI data was collected from thirty-two patients with geriatric depression and thirty-two age-matched never-depressed controls. We compared the resting-state activities between the two groups in the default-mode, central executive, attention, salience, and affective networks as well as correlations among these networks. The depression group showed stronger activity than the controls in an affective network, specifically within the orbitofrontal region. However, unlike the never-depressed controls, geriatric depression group lacked synchronized/antisynchronized activity between the affective network and the other networks. Those depressed patients with lower executive function has greater synchronization between the salience network with the executive and affective networks. Our results demonstrate the effectiveness of the between-network analyses in examining neural models for geriatric depression. PMID:26180795

Altered Synchronizations among Neural Networks in Geriatric Depression.

PubMed

Wang, Lihong; Chou, Ying-Hui; Potter, Guy G; Steffens, David C

2015-01-01

Although major depression has been considered as a manifestation of discoordinated activity between affective and cognitive neural networks, only a few studies have examined the relationships among neural networks directly. Because of the known disconnection theory, geriatric depression could be a useful model in studying the interactions among different networks. In the present study, using independent component analysis to identify intrinsically connected neural networks, we investigated the alterations in synchronizations among neural networks in geriatric depression to better understand the underlying neural mechanisms. Resting-state fMRI data was collected from thirty-two patients with geriatric depression and thirty-two age-matched never-depressed controls. We compared the resting-state activities between the two groups in the default-mode, central executive, attention, salience, and affective networks as well as correlations among these networks. The depression group showed stronger activity than the controls in an affective network, specifically within the orbitofrontal region. However, unlike the never-depressed controls, geriatric depression group lacked synchronized/antisynchronized activity between the affective network and the other networks. Those depressed patients with lower executive function has greater synchronization between the salience network with the executive and affective networks. Our results demonstrate the effectiveness of the between-network analyses in examining neural models for geriatric depression.

A consensual neural network

NASA Technical Reports Server (NTRS)

Benediktsson, J. A.; Ersoy, O. K.; Swain, P. H.

1991-01-01

A neural network architecture called a consensual neural network (CNN) is proposed for the classification of data from multiple sources. Its relation to hierarchical and ensemble neural networks is discussed. CNN is based on the statistical consensus theory and uses nonlinearly transformed input data. The input data are transformed several times, and the different transformed data are applied as if they were independent inputs. The independent inputs are classified using stage neural networks and outputs from the stage networks are then weighted and combined to make a decision. Experimental results based on remote-sensing data and geographic data are given.

Neural-Network Simulator

NASA Technical Reports Server (NTRS)

Mitchell, Paul H.

1991-01-01

F77NNS (FORTRAN 77 Neural Network Simulator) computer program simulates popular back-error-propagation neural network. Designed to take advantage of vectorization when used on computers having this capability, also used on any computer equipped with ANSI-77 FORTRAN Compiler. Problems involving matching of patterns or mathematical modeling of systems fit class of problems F77NNS designed to solve. Program has restart capability so neural network solved in stages suitable to user's resources and desires. Enables user to customize patterns of connections between layers of network. Size of neural network F77NNS applied to limited only by amount of random-access memory available to user.

Feedback modulation of neural network synchrony and seizure susceptibility by Mdm2-p53-Nedd4-2 signaling.

PubMed

Jewett, Kathryn A; Christian, Catherine A; Bacos, Jonathan T; Lee, Kwan Young; Zhu, Jiuhe; Tsai, Nien-Pei

2016-03-22

Neural network synchrony is a critical factor in regulating information transmission through the nervous system. Improperly regulated neural network synchrony is implicated in pathophysiological conditions such as epilepsy. Despite the awareness of its importance, the molecular signaling underlying the regulation of neural network synchrony, especially after stimulation, remains largely unknown. In this study, we show that elevation of neuronal activity by the GABA(A) receptor antagonist, Picrotoxin, increases neural network synchrony in primary mouse cortical neuron cultures. The elevation of neuronal activity triggers Mdm2-dependent degradation of the tumor suppressor p53. We show here that blocking the degradation of p53 further enhances Picrotoxin-induced neural network synchrony, while promoting the inhibition of p53 with a p53 inhibitor reduces Picrotoxin-induced neural network synchrony. These data suggest that Mdm2-p53 signaling mediates a feedback mechanism to fine-tune neural network synchrony after activity stimulation. Furthermore, genetically reducing the expression of a direct target gene of p53, Nedd4-2, elevates neural network synchrony basally and occludes the effect of Picrotoxin. Finally, using a kainic acid-induced seizure model in mice, we show that alterations of Mdm2-p53-Nedd4-2 signaling affect seizure susceptibility. Together, our findings elucidate a critical role of Mdm2-p53-Nedd4-2 signaling underlying the regulation of neural network synchrony and seizure susceptibility and reveal potential therapeutic targets for hyperexcitability-associated neurological disorders.

Neural network-based model reference adaptive control system.

PubMed

Patino, H D; Liu, D

2000-01-01

In this paper, an approach to model reference adaptive control based on neural networks is proposed and analyzed for a class of first-order continuous-time nonlinear dynamical systems. The controller structure can employ either a radial basis function network or a feedforward neural network to compensate adaptively the nonlinearities in the plant. A stable controller-parameter adjustment mechanism, which is determined using the Lyapunov theory, is constructed using a sigma-modification-type updating law. The evaluation of control error in terms of the neural network learning error is performed. That is, the control error converges asymptotically to a neighborhood of zero, whose size is evaluated and depends on the approximation error of the neural network. In the design and analysis of neural network-based control systems, it is important to take into account the neural network learning error and its influence on the control error of the plant. Simulation results showing the feasibility and performance of the proposed approach are given.

Medical subdomain classification of clinical notes using a machine learning-based natural language processing approach.

PubMed

Weng, Wei-Hung; Wagholikar, Kavishwar B; McCray, Alexa T; Szolovits, Peter; Chueh, Henry C

2017-12-01

The medical subdomain of a clinical note, such as cardiology or neurology, is useful content-derived metadata for developing machine learning downstream applications. To classify the medical subdomain of a note accurately, we have constructed a machine learning-based natural language processing (NLP) pipeline and developed medical subdomain classifiers based on the content of the note. We constructed the pipeline using the clinical NLP system, clinical Text Analysis and Knowledge Extraction System (cTAKES), the Unified Medical Language System (UMLS) Metathesaurus, Semantic Network, and learning algorithms to extract features from two datasets - clinical notes from Integrating Data for Analysis, Anonymization, and Sharing (iDASH) data repository (n = 431) and Massachusetts General Hospital (MGH) (n = 91,237), and built medical subdomain classifiers with different combinations of data representation methods and supervised learning algorithms. We evaluated the performance of classifiers and their portability across the two datasets. The convolutional recurrent neural network with neural word embeddings trained-medical subdomain classifier yielded the best performance measurement on iDASH and MGH datasets with area under receiver operating characteristic curve (AUC) of 0.975 and 0.991, and F1 scores of 0.845 and 0.870, respectively. Considering better clinical interpretability, linear support vector machine-trained medical subdomain classifier using hybrid bag-of-words and clinically relevant UMLS concepts as the feature representation, with term frequency-inverse document frequency (tf-idf)-weighting, outperformed other shallow learning classifiers on iDASH and MGH datasets with AUC of 0.957 and 0.964, and F1 scores of 0.932 and 0.934 respectively. We trained classifiers on one dataset, applied to the other dataset and yielded the threshold of F1 score of 0.7 in classifiers for half of the medical subdomains we studied. Our study shows that a supervised learning-based NLP approach is useful to develop medical subdomain classifiers. The deep learning algorithm with distributed word representation yields better performance yet shallow learning algorithms with the word and concept representation achieves comparable performance with better clinical interpretability. Portable classifiers may also be used across datasets from different institutions.

Space-Time Neural Networks

NASA Technical Reports Server (NTRS)

Villarreal, James A.; Shelton, Robert O.

1992-01-01

Concept of space-time neural network affords distributed temporal memory enabling such network to model complicated dynamical systems mathematically and to recognize temporally varying spatial patterns. Digital filters replace synaptic-connection weights of conventional back-error-propagation neural network.

Effects of bursting dynamic features on the generation of multi-clustered structure of neural network with symmetric spike-timing-dependent plasticity learning rule.

PubMed

Liu, Hui; Song, Yongduan; Xue, Fangzheng; Li, Xiumin

2015-11-01

In this paper, the generation of multi-clustered structure of self-organized neural network with different neuronal firing patterns, i.e., bursting or spiking, has been investigated. The initially all-to-all-connected spiking neural network or bursting neural network can be self-organized into clustered structure through the symmetric spike-timing-dependent plasticity learning for both bursting and spiking neurons. However, the time consumption of this clustering procedure of the burst-based self-organized neural network (BSON) is much shorter than the spike-based self-organized neural network (SSON). Our results show that the BSON network has more obvious small-world properties, i.e., higher clustering coefficient and smaller shortest path length than the SSON network. Also, the results of larger structure entropy and activity entropy of the BSON network demonstrate that this network has higher topological complexity and dynamical diversity, which benefits for enhancing information transmission of neural circuits. Hence, we conclude that the burst firing can significantly enhance the efficiency of clustering procedure and the emergent clustered structure renders the whole network more synchronous and therefore more sensitive to weak input. This result is further confirmed from its improved performance on stochastic resonance. Therefore, we believe that the multi-clustered neural network which self-organized from the bursting dynamics has high efficiency in information processing.

Effects of bursting dynamic features on the generation of multi-clustered structure of neural network with symmetric spike-timing-dependent plasticity learning rule

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

Liu, Hui; Song, Yongduan; Xue, Fangzheng

In this paper, the generation of multi-clustered structure of self-organized neural network with different neuronal firing patterns, i.e., bursting or spiking, has been investigated. The initially all-to-all-connected spiking neural network or bursting neural network can be self-organized into clustered structure through the symmetric spike-timing-dependent plasticity learning for both bursting and spiking neurons. However, the time consumption of this clustering procedure of the burst-based self-organized neural network (BSON) is much shorter than the spike-based self-organized neural network (SSON). Our results show that the BSON network has more obvious small-world properties, i.e., higher clustering coefficient and smaller shortest path length than themore » SSON network. Also, the results of larger structure entropy and activity entropy of the BSON network demonstrate that this network has higher topological complexity and dynamical diversity, which benefits for enhancing information transmission of neural circuits. Hence, we conclude that the burst firing can significantly enhance the efficiency of clustering procedure and the emergent clustered structure renders the whole network more synchronous and therefore more sensitive to weak input. This result is further confirmed from its improved performance on stochastic resonance. Therefore, we believe that the multi-clustered neural network which self-organized from the bursting dynamics has high efficiency in information processing.« less

Financial Time Series Prediction Using Spiking Neural Networks

PubMed Central

Reid, David; Hussain, Abir Jaafar; Tawfik, Hissam

2014-01-01

In this paper a novel application of a particular type of spiking neural network, a Polychronous Spiking Network, was used for financial time series prediction. It is argued that the inherent temporal capabilities of this type of network are suited to non-stationary data such as this. The performance of the spiking neural network was benchmarked against three systems: two “traditional”, rate-encoded, neural networks; a Multi-Layer Perceptron neural network and a Dynamic Ridge Polynomial neural network, and a standard Linear Predictor Coefficients model. For this comparison three non-stationary and noisy time series were used: IBM stock data; US/Euro exchange rate data, and the price of Brent crude oil. The experiments demonstrated favourable prediction results for the Spiking Neural Network in terms of Annualised Return and prediction error for 5-Step ahead predictions. These results were also supported by other relevant metrics such as Maximum Drawdown and Signal-To-Noise ratio. This work demonstrated the applicability of the Polychronous Spiking Network to financial data forecasting and this in turn indicates the potential of using such networks over traditional systems in difficult to manage non-stationary environments. PMID:25170618

Qualitative analysis of Cohen-Grossberg neural networks with multiple delays

NASA Astrophysics Data System (ADS)

Ye, Hui; Michel, Anthony N.; Wang, Kaining

1995-03-01

It is well known that a class of artificial neural networks with symmetric interconnections and without transmission delays, known as Cohen-Grossberg neural networks, possesses global stability (i.e., all trajectories tend to some equilibrium). We demonstrate in the present paper that many of the qualitative properties of Cohen-Grossberg networks will not be affected by the introduction of sufficiently small delays. Specifically, we establish some bound conditions for the time delays under which a given Cohen-Grossberg network with multiple delays is globally stable and possesses the same asymptotically stable equilibria as the corresponding network without delays. An effective method of determining the asymptotic stability of an equilibrium of a Cohen-Grossberg network with multiple delays is also presented. The present results are motivated by some of the authors earlier work [Phys. Rev. E 50, 4206 (1994)] and by some of the work of Marcus and Westervelt [Phys. Rev. A 39, 347 (1989)]. These works address qualitative analyses of Hopfield neural networks with one time delay. The present work generalizes these results to Cohen-Grossberg neural networks with multiple time delays. Hopfield neural networks constitute special cases of Cohen-Grossberg neural networks.

Dynamic Neural Networks Supporting Memory Retrieval

PubMed Central

St. Jacques, Peggy L.; Kragel, Philip A.; Rubin, David C.

2011-01-01

How do separate neural networks interact to support complex cognitive processes such as remembrance of the personal past? Autobiographical memory (AM) retrieval recruits a consistent pattern of activation that potentially comprises multiple neural networks. However, it is unclear how such large-scale neural networks interact and are modulated by properties of the memory retrieval process. In the present functional MRI (fMRI) study, we combined independent component analysis (ICA) and dynamic causal modeling (DCM) to understand the neural networks supporting AM retrieval. ICA revealed four task-related components consistent with the previous literature: 1) Medial Prefrontal Cortex (PFC) Network, associated with self-referential processes, 2) Medial Temporal Lobe (MTL) Network, associated with memory, 3) Frontoparietal Network, associated with strategic search, and 4) Cingulooperculum Network, associated with goal maintenance. DCM analysis revealed that the medial PFC network drove activation within the system, consistent with the importance of this network to AM retrieval. Additionally, memory accessibility and recollection uniquely altered connectivity between these neural networks. Recollection modulated the influence of the medial PFC on the MTL network during elaboration, suggesting that greater connectivity among subsystems of the default network supports greater re-experience. In contrast, memory accessibility modulated the influence of frontoparietal and MTL networks on the medial PFC network, suggesting that ease of retrieval involves greater fluency among the multiple networks contributing to AM. These results show the integration between neural networks supporting AM retrieval and the modulation of network connectivity by behavior. PMID:21550407

Coherence resonance in bursting neural networks

NASA Astrophysics Data System (ADS)

Kim, June Hoan; Lee, Ho Jun; Min, Cheol Hong; Lee, Kyoung J.

2015-10-01

Synchronized neural bursts are one of the most noticeable dynamic features of neural networks, being essential for various phenomena in neuroscience, yet their complex dynamics are not well understood. With extrinsic electrical and optical manipulations on cultured neural networks, we demonstrate that the regularity (or randomness) of burst sequences is in many cases determined by a (few) low-dimensional attractor(s) working under strong neural noise. Moreover, there is an optimal level of noise strength at which the regularity of the interburst interval sequence becomes maximal—a phenomenon of coherence resonance. The experimental observations are successfully reproduced through computer simulations on a well-established neural network model, suggesting that the same phenomena may occur in many in vivo as well as in vitro neural networks.

Classification of Respiratory Sounds by Using An Artificial Neural Network

DTIC Science & Technology

2001-10-28

CLASSIFICATION OF RESPIRATORY SOUNDS BY USING AN ARTIFICIAL NEURAL NETWORK M.C. Sezgin, Z. Dokur, T. Ölmez, M. Korürek Department of Electronics and...successfully classified by the GAL network. Keywords-Respiratory Sounds, Classification of Biomedical Signals, Artificial Neural Network . I. INTRODUCTION...process, feature extraction, and classification by the artificial neural network . At first, the RS signal obtained from a real-time measurement equipment is

Instrumentation for Scientific Computing in Neural Networks, Information Science, Artificial Intelligence, and Applied Mathematics.

DTIC Science & Technology

1987-10-01

include Security Classification) Instrumentation for scientific computing in neural networks, information science, artificial intelligence, and...instrumentation grant to purchase equipment for support of research in neural networks, information science, artificail intellignece , and applied mathematics...in Neural Networks, Information Science, Artificial Intelligence, and Applied Mathematics Contract AFOSR 86-0282 Principal Investigator: Stephen

A neural net approach to space vehicle guidance

NASA Technical Reports Server (NTRS)

Caglayan, Alper K.; Allen, Scott M.

1990-01-01

The space vehicle guidance problem is formulated using a neural network approach, and the appropriate neural net architecture for modeling optimum guidance trajectories is investigated. In particular, an investigation is made of the incorporation of prior knowledge about the characteristics of the optimal guidance solution into the neural network architecture. The online classification performance of the developed network is demonstrated using a synthesized network trained with a database of optimum guidance trajectories. Such a neural-network-based guidance approach can readily adapt to environment uncertainties such as those encountered by an AOTV during atmospheric maneuvers.

Neural network and its application to CT imaging

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

Nikravesh, M.; Kovscek, A.R.; Patzek, T.W.

We present an integrated approach to imaging the progress of air displacement by spontaneous imbibition of oil into sandstone. We combine Computerized Tomography (CT) scanning and neural network image processing. The main aspects of our approach are (I) visualization of the distribution of oil and air saturation by CT, (II) interpretation of CT scans using neural networks, and (III) reconstruction of 3-D images of oil saturation from the CT scans with a neural network model. Excellent agreement between the actual images and the neural network predictions is found.

Electronic neural networks for global optimization

NASA Technical Reports Server (NTRS)

Thakoor, A. P.; Moopenn, A. W.; Eberhardt, S.

1990-01-01

An electronic neural network with feedback architecture, implemented in analog custom VLSI is described. Its application to problems of global optimization for dynamic assignment is discussed. The convergence properties of the neural network hardware are compared with computer simulation results. The neural network's ability to provide optimal or near optimal solutions within only a few neuron time constants, a speed enhancement of several orders of magnitude over conventional search methods, is demonstrated. The effect of noise on the circuit dynamics and the convergence behavior of the neural network hardware is also examined.

Quantitative analysis of volatile organic compounds using ion mobility spectra and cascade correlation neural networks

NASA Technical Reports Server (NTRS)

Harrington, Peter DEB.; Zheng, Peng

1995-01-01

Ion Mobility Spectrometry (IMS) is a powerful technique for trace organic analysis in the gas phase. Quantitative measurements are difficult, because IMS has a limited linear range. Factors that may affect the instrument response are pressure, temperature, and humidity. Nonlinear calibration methods, such as neural networks, may be ideally suited for IMS. Neural networks have the capability of modeling complex systems. Many neural networks suffer from long training times and overfitting. Cascade correlation neural networks train at very fast rates. They also build their own topology, that is a number of layers and number of units in each layer. By controlling the decay parameter in training neural networks, reproducible and general models may be obtained.

Newly developed double neural network concept for reliable fast plasma position control

NASA Astrophysics Data System (ADS)

Jeon, Young-Mu; Na, Yong-Su; Kim, Myung-Rak; Hwang, Y. S.

2001-01-01

Neural network is considered as a parameter estimation tool in plasma controls for next generation tokamak such as ITER. The neural network has been reported to be so accurate and fast for plasma equilibrium identification that it may be applied to the control of complex tokamak plasmas. For this application, the reliability of the conventional neural network needs to be improved. In this study, a new idea of double neural network is developed to achieve this. The new idea has been applied to simple plasma position identification of KSTAR tokamak for feasibility test. Characteristics of the concept show higher reliability and fault tolerance even in severe faulty conditions, which may make neural network applicable to plasma control reliably and widely in future tokamaks.

Rule extraction from minimal neural networks for credit card screening.

PubMed

Setiono, Rudy; Baesens, Bart; Mues, Christophe

2011-08-01

While feedforward neural networks have been widely accepted as effective tools for solving classification problems, the issue of finding the best network architecture remains unresolved, particularly so in real-world problem settings. We address this issue in the context of credit card screening, where it is important to not only find a neural network with good predictive performance but also one that facilitates a clear explanation of how it produces its predictions. We show that minimal neural networks with as few as one hidden unit provide good predictive accuracy, while having the added advantage of making it easier to generate concise and comprehensible classification rules for the user. To further reduce model size, a novel approach is suggested in which network connections from the input units to this hidden unit are removed by a very straightaway pruning procedure. In terms of predictive accuracy, both the minimized neural networks and the rule sets generated from them are shown to compare favorably with other neural network based classifiers. The rules generated from the minimized neural networks are concise and thus easier to validate in a real-life setting.

An improved wavelet neural network medical image segmentation algorithm with combined maximum entropy

NASA Astrophysics Data System (ADS)

Hu, Xiaoqian; Tao, Jinxu; Ye, Zhongfu; Qiu, Bensheng; Xu, Jinzhang

2018-05-01

In order to solve the problem of medical image segmentation, a wavelet neural network medical image segmentation algorithm based on combined maximum entropy criterion is proposed. Firstly, we use bee colony algorithm to optimize the network parameters of wavelet neural network, get the parameters of network structure, initial weights and threshold values, and so on, we can quickly converge to higher precision when training, and avoid to falling into relative extremum; then the optimal number of iterations is obtained by calculating the maximum entropy of the segmented image, so as to achieve the automatic and accurate segmentation effect. Medical image segmentation experiments show that the proposed algorithm can reduce sample training time effectively and improve convergence precision, and segmentation effect is more accurate and effective than traditional BP neural network (back propagation neural network : a multilayer feed forward neural network which trained according to the error backward propagation algorithm.

Knowledge extraction from evolving spiking neural networks with rank order population coding.

PubMed

Soltic, Snjezana; Kasabov, Nikola

2010-12-01

This paper demonstrates how knowledge can be extracted from evolving spiking neural networks with rank order population coding. Knowledge discovery is a very important feature of intelligent systems. Yet, a disproportionally small amount of research is centered on the issue of knowledge extraction from spiking neural networks which are considered to be the third generation of artificial neural networks. The lack of knowledge representation compatibility is becoming a major detriment to end users of these networks. We show that a high-level knowledge can be obtained from evolving spiking neural networks. More specifically, we propose a method for fuzzy rule extraction from an evolving spiking network with rank order population coding. The proposed method was used for knowledge discovery on two benchmark taste recognition problems where the knowledge learnt by an evolving spiking neural network was extracted in the form of zero-order Takagi-Sugeno fuzzy IF-THEN rules.

Integrative gene network construction to analyze cancer recurrence using semi-supervised learning.

PubMed

Park, Chihyun; Ahn, Jaegyoon; Kim, Hyunjin; Park, Sanghyun

2014-01-01

The prognosis of cancer recurrence is an important research area in bioinformatics and is challenging due to the small sample sizes compared to the vast number of genes. There have been several attempts to predict cancer recurrence. Most studies employed a supervised approach, which uses only a few labeled samples. Semi-supervised learning can be a great alternative to solve this problem. There have been few attempts based on manifold assumptions to reveal the detailed roles of identified cancer genes in recurrence. In order to predict cancer recurrence, we proposed a novel semi-supervised learning algorithm based on a graph regularization approach. We transformed the gene expression data into a graph structure for semi-supervised learning and integrated protein interaction data with the gene expression data to select functionally-related gene pairs. Then, we predicted the recurrence of cancer by applying a regularization approach to the constructed graph containing both labeled and unlabeled nodes. The average improvement rate of accuracy for three different cancer datasets was 24.9% compared to existing supervised and semi-supervised methods. We performed functional enrichment on the gene networks used for learning. We identified that those gene networks are significantly associated with cancer-recurrence-related biological functions. Our algorithm was developed with standard C++ and is available in Linux and MS Windows formats in the STL library. The executable program is freely available at: http://embio.yonsei.ac.kr/~Park/ssl.php.

A Customized Attention-Based Long Short-Term Memory Network for Distant Supervised Relation Extraction.

PubMed

He, Dengchao; Zhang, Hongjun; Hao, Wenning; Zhang, Rui; Cheng, Kai

2017-07-01

Distant supervision, a widely applied approach in the field of relation extraction can automatically generate large amounts of labeled training corpus with minimal manual effort. However, the labeled training corpus may have many false-positive data, which would hurt the performance of relation extraction. Moreover, in traditional feature-based distant supervised approaches, extraction models adopt human design features with natural language processing. It may also cause poor performance. To address these two shortcomings, we propose a customized attention-based long short-term memory network. Our approach adopts word-level attention to achieve better data representation for relation extraction without manually designed features to perform distant supervision instead of fully supervised relation extraction, and it utilizes instance-level attention to tackle the problem of false-positive data. Experimental results demonstrate that our proposed approach is effective and achieves better performance than traditional methods.

Active semi-supervised learning method with hybrid deep belief networks.

PubMed

Zhou, Shusen; Chen, Qingcai; Wang, Xiaolong

2014-01-01

In this paper, we develop a novel semi-supervised learning algorithm called active hybrid deep belief networks (AHD), to address the semi-supervised sentiment classification problem with deep learning. First, we construct the previous several hidden layers using restricted Boltzmann machines (RBM), which can reduce the dimension and abstract the information of the reviews quickly. Second, we construct the following hidden layers using convolutional restricted Boltzmann machines (CRBM), which can abstract the information of reviews effectively. Third, the constructed deep architecture is fine-tuned by gradient-descent based supervised learning with an exponential loss function. Finally, active learning method is combined based on the proposed deep architecture. We did several experiments on five sentiment classification datasets, and show that AHD is competitive with previous semi-supervised learning algorithm. Experiments are also conducted to verify the effectiveness of our proposed method with different number of labeled reviews and unlabeled reviews respectively.

Adaptive neural network motion control of manipulators with experimental evaluations.

PubMed

Puga-Guzmán, S; Moreno-Valenzuela, J; Santibáñez, V

2014-01-01

A nonlinear proportional-derivative controller plus adaptive neuronal network compensation is proposed. With the aim of estimating the desired torque, a two-layer neural network is used. Then, adaptation laws for the neural network weights are derived. Asymptotic convergence of the position and velocity tracking errors is proven, while the neural network weights are shown to be uniformly bounded. The proposed scheme has been experimentally validated in real time. These experimental evaluations were carried in two different mechanical systems: a horizontal two degrees-of-freedom robot and a vertical one degree-of-freedom arm which is affected by the gravitational force. In each one of the two experimental set-ups, the proposed scheme was implemented without and with adaptive neural network compensation. Experimental results confirmed the tracking accuracy of the proposed adaptive neural network-based controller.

Adaptive Neural Network Motion Control of Manipulators with Experimental Evaluations

PubMed Central

Puga-Guzmán, S.; Moreno-Valenzuela, J.; Santibáñez, V.

2014-01-01

A nonlinear proportional-derivative controller plus adaptive neuronal network compensation is proposed. With the aim of estimating the desired torque, a two-layer neural network is used. Then, adaptation laws for the neural network weights are derived. Asymptotic convergence of the position and velocity tracking errors is proven, while the neural network weights are shown to be uniformly bounded. The proposed scheme has been experimentally validated in real time. These experimental evaluations were carried in two different mechanical systems: a horizontal two degrees-of-freedom robot and a vertical one degree-of-freedom arm which is affected by the gravitational force. In each one of the two experimental set-ups, the proposed scheme was implemented without and with adaptive neural network compensation. Experimental results confirmed the tracking accuracy of the proposed adaptive neural network-based controller. PMID:24574910

Research on image retrieval using deep convolutional neural network combining L1 regularization and PRelu activation function

NASA Astrophysics Data System (ADS)

QingJie, Wei; WenBin, Wang

2017-06-01

In this paper, the image retrieval using deep convolutional neural network combined with regularization and PRelu activation function is studied, and improves image retrieval accuracy. Deep convolutional neural network can not only simulate the process of human brain to receive and transmit information, but also contains a convolution operation, which is very suitable for processing images. Using deep convolutional neural network is better than direct extraction of image visual features for image retrieval. However, the structure of deep convolutional neural network is complex, and it is easy to over-fitting and reduces the accuracy of image retrieval. In this paper, we combine L1 regularization and PRelu activation function to construct a deep convolutional neural network to prevent over-fitting of the network and improve the accuracy of image retrieval

A Cognitive Neural Architecture Able to Learn and Communicate through Natural Language.

PubMed

Golosio, Bruno; Cangelosi, Angelo; Gamotina, Olesya; Masala, Giovanni Luca

2015-01-01

Communicative interactions involve a kind of procedural knowledge that is used by the human brain for processing verbal and nonverbal inputs and for language production. Although considerable work has been done on modeling human language abilities, it has been difficult to bring them together to a comprehensive tabula rasa system compatible with current knowledge of how verbal information is processed in the brain. This work presents a cognitive system, entirely based on a large-scale neural architecture, which was developed to shed light on the procedural knowledge involved in language elaboration. The main component of this system is the central executive, which is a supervising system that coordinates the other components of the working memory. In our model, the central executive is a neural network that takes as input the neural activation states of the short-term memory and yields as output mental actions, which control the flow of information among the working memory components through neural gating mechanisms. The proposed system is capable of learning to communicate through natural language starting from tabula rasa, without any a priori knowledge of the structure of phrases, meaning of words, role of the different classes of words, only by interacting with a human through a text-based interface, using an open-ended incremental learning process. It is able to learn nouns, verbs, adjectives, pronouns and other word classes, and to use them in expressive language. The model was validated on a corpus of 1587 input sentences, based on literature on early language assessment, at the level of about 4-years old child, and produced 521 output sentences, expressing a broad range of language processing functionalities.

Program Helps Simulate Neural Networks

NASA Technical Reports Server (NTRS)

Villarreal, James; Mcintire, Gary

1993-01-01

Neural Network Environment on Transputer System (NNETS) computer program provides users high degree of flexibility in creating and manipulating wide variety of neural-network topologies at processing speeds not found in conventional computing environments. Supports back-propagation and back-propagation-related algorithms. Back-propagation algorithm used is implementation of Rumelhart's generalized delta rule. NNETS developed on INMOS Transputer(R). Predefines back-propagation network, Jordan network, and reinforcement network to assist users in learning and defining own networks. Also enables users to configure other neural-network paradigms from NNETS basic architecture. Small portion of software written in OCCAM(R) language.

Establishing a Dynamic Self-Adaptation Learning Algorithm of the BP Neural Network and Its Applications

NASA Astrophysics Data System (ADS)

Li, Xiaofeng; Xiang, Suying; Zhu, Pengfei; Wu, Min

2015-12-01

In order to avoid the inherent deficiencies of the traditional BP neural network, such as slow convergence speed, that easily leading to local minima, poor generalization ability and difficulty in determining the network structure, the dynamic self-adaptive learning algorithm of the BP neural network is put forward to improve the function of the BP neural network. The new algorithm combines the merit of principal component analysis, particle swarm optimization, correlation analysis and self-adaptive model, hence can effectively solve the problems of selecting structural parameters, initial connection weights and thresholds and learning rates of the BP neural network. This new algorithm not only reduces the human intervention, optimizes the topological structures of BP neural networks and improves the network generalization ability, but also accelerates the convergence speed of a network, avoids trapping into local minima, and enhances network adaptation ability and prediction ability. The dynamic self-adaptive learning algorithm of the BP neural network is used to forecast the total retail sale of consumer goods of Sichuan Province, China. Empirical results indicate that the new algorithm is superior to the traditional BP network algorithm in predicting accuracy and time consumption, which shows the feasibility and effectiveness of the new algorithm.

Neural net target-tracking system using structured laser patterns

NASA Astrophysics Data System (ADS)

Cho, Jae-Wan; Lee, Yong-Bum; Lee, Nam-Ho; Park, Soon-Yong; Lee, Jongmin; Choi, Gapchu; Baek, Sunghyun; Park, Dong-Sun

1996-06-01

In this paper, we describe a robot endeffector tracking system using sensory information from recently-announced structured pattern laser diodes, which can generate images with several different types of structured pattern. The neural network approach is employed to recognize the robot endeffector covering the situation of three types of motion: translation, scaling and rotation. Features for the neural network to detect the position of the endeffector are extracted from the preprocessed images. Artificial neural networks are used to store models and to match with unknown input features recognizing the position of the robot endeffector. Since a minimal number of samples are used for different directions of the robot endeffector in the system, an artificial neural network with the generalization capability can be utilized for unknown input features. A feedforward neural network with the generalization capability can be utilized for unknown input features. A feedforward neural network trained with the back propagation learning is used to detect the position of the robot endeffector. Another feedforward neural network module is used to estimate the motion from a sequence of images and to control movements of the robot endeffector. COmbining the tow neural networks for recognizing the robot endeffector and estimating the motion with the preprocessing stage, the whole system keeps tracking of the robot endeffector effectively.

Chaotic simulated annealing by a neural network with a variable delay: design and application.

PubMed

Chen, Shyan-Shiou

2011-10-01

In this paper, we have three goals: the first is to delineate the advantages of a variably delayed system, the second is to find a more intuitive Lyapunov function for a delayed neural network, and the third is to design a delayed neural network for a quadratic cost function. For delayed neural networks, most researchers construct a Lyapunov function based on the linear matrix inequality (LMI) approach. However, that approach is not intuitive. We provide a alternative candidate Lyapunov function for a delayed neural network. On the other hand, if we are first given a quadratic cost function, we can construct a delayed neural network by suitably dividing the second-order term into two parts: a self-feedback connection weight and a delayed connection weight. To demonstrate the advantage of a variably delayed neural network, we propose a transiently chaotic neural network with variable delay and show numerically that the model should possess a better searching ability than Chen-Aihara's model, Wang's model, and Zhao's model. We discuss both the chaotic and the convergent phases. During the chaotic phase, we simply present bifurcation diagrams for a single neuron with a constant delay and with a variable delay. We show that the variably delayed model possesses the stochastic property and chaotic wandering. During the convergent phase, we not only provide a novel Lyapunov function for neural networks with a delay (the Lyapunov function is independent of the LMI approach) but also establish a correlation between the Lyapunov function for a delayed neural network and an objective function for the traveling salesman problem. © 2011 IEEE

Deep neural networks for direct, featureless learning through observation: The case of two-dimensional spin models

NASA Astrophysics Data System (ADS)

Mills, Kyle; Tamblyn, Isaac

2018-03-01

We demonstrate the capability of a convolutional deep neural network in predicting the nearest-neighbor energy of the 4 ×4 Ising model. Using its success at this task, we motivate the study of the larger 8 ×8 Ising model, showing that the deep neural network can learn the nearest-neighbor Ising Hamiltonian after only seeing a vanishingly small fraction of configuration space. Additionally, we show that the neural network has learned both the energy and magnetization operators with sufficient accuracy to replicate the low-temperature Ising phase transition. We then demonstrate the ability of the neural network to learn other spin models, teaching the convolutional deep neural network to accurately predict the long-range interaction of a screened Coulomb Hamiltonian, a sinusoidally attenuated screened Coulomb Hamiltonian, and a modified Potts model Hamiltonian. In the case of the long-range interaction, we demonstrate the ability of the neural network to recover the phase transition with equivalent accuracy to the numerically exact method. Furthermore, in the case of the long-range interaction, the benefits of the neural network become apparent; it is able to make predictions with a high degree of accuracy, and do so 1600 times faster than a CUDA-optimized exact calculation. Additionally, we demonstrate how the neural network succeeds at these tasks by looking at the weights learned in a simplified demonstration.

Classification of boreal forest by satellite and inventory data using neural network approach

NASA Astrophysics Data System (ADS)

Romanov, A. A.

2012-12-01

The main objective of this research was to develop methodology for boreal (Siberian Taiga) land cover classification in a high accuracy level. The study area covers the territories of Central Siberian several parts along the Yenisei River (60-62 degrees North Latitude): the right bank includes mixed forest and dark taiga, the left - pine forests; so were taken as a high heterogeneity and statistically equal surfaces concerning spectral characteristics. Two main types of data were used: time series of middle spatial resolution satellite images (Landsat 5, 7 and SPOT4) and inventory datasets from the nature fieldworks (used for training samples sets preparation). Method of collecting field datasets included a short botany description (type/species of vegetation, density, compactness of the crowns, individual height and max/min diameters representative of each type, surface altitude of the plot), at the same time the geometric characteristic of each training sample unit corresponded to the spatial resolution of satellite images and geo-referenced (prepared datasets both of the preliminary processing and verification). The network of test plots was planned as irregular and determined by the landscape oriented approach. The main focus of the thematic data processing has been allocated for the use of neural networks (fuzzy logic inc.); therefore, the results of field studies have been converting input parameter of type / species of vegetation cover of each unit and the degree of variability. Proposed approach involves the processing of time series separately for each image mainly for the verification: shooting parameters taken into consideration (time, albedo) and thus expected to assess the quality of mapping. So the input variables for the networks were sensor bands, surface altitude, solar angels and land surface temperature (for a few experiments); also given attention to the formation of the formula class on the basis of statistical pre-processing of results of field research (prevalence type). Besides some statistical methods of supervised classification has been used (minimal distance, maximum likelihood, Mahalanobis). During the study received various types of neural classifiers suitable for the mapping, and even for the high heterogenic areas neural network approach has shown better results in precision despite the validity of the assumption of Gaussian distribution (Table). Experimentally chosen optimum network structure consisting of three layers of ten neuron in each, but it should be clarified that such configuration requires larges computational resources in comparison the statistical methods presented above; necessary to increase the number of iteration in network learning process for RMS errors minimization. It should also be emphasized that the key issues of accuracy estimation of the classification results is lack of completeness of the training sets, this is especially true with summer image processing of mixed forest. However seems that proposed methodology can be used also for measure local dynamic of boreal land surface by the type of vegetation.Comparison of classification accuracyt;

A Multiagent-based Intrusion Detection System with the Support of Multi-Class Supervised Classification

NASA Astrophysics Data System (ADS)

Shyu, Mei-Ling; Sainani, Varsha

The increasing number of network security related incidents have made it necessary for the organizations to actively protect their sensitive data with network intrusion detection systems (IDSs). IDSs are expected to analyze a large volume of data while not placing a significantly added load on the monitoring systems and networks. This requires good data mining strategies which take less time and give accurate results. In this study, a novel data mining assisted multiagent-based intrusion detection system (DMAS-IDS) is proposed, particularly with the support of multiclass supervised classification. These agents can detect and take predefined actions against malicious activities, and data mining techniques can help detect them. Our proposed DMAS-IDS shows superior performance compared to central sniffing IDS techniques, and saves network resources compared to other distributed IDS with mobile agents that activate too many sniffers causing bottlenecks in the network. This is one of the major motivations to use a distributed model based on multiagent platform along with a supervised classification technique.

Tensor Basis Neural Network v. 1.0 (beta)

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

Ling, Julia; Templeton, Jeremy

This software package can be used to build, train, and test a neural network machine learning model. The neural network architecture is specifically designed to embed tensor invariance properties by enforcing that the model predictions sit on an invariant tensor basis. This neural network architecture can be used in developing constitutive models for applications such as turbulence modeling, materials science, and electromagnetism.

A renaissance of neural networks in drug discovery.

PubMed

Baskin, Igor I; Winkler, David; Tetko, Igor V

2016-08-01

Neural networks are becoming a very popular method for solving machine learning and artificial intelligence problems. The variety of neural network types and their application to drug discovery requires expert knowledge to choose the most appropriate approach. In this review, the authors discuss traditional and newly emerging neural network approaches to drug discovery. Their focus is on backpropagation neural networks and their variants, self-organizing maps and associated methods, and a relatively new technique, deep learning. The most important technical issues are discussed including overfitting and its prevention through regularization, ensemble and multitask modeling, model interpretation, and estimation of applicability domain. Different aspects of using neural networks in drug discovery are considered: building structure-activity models with respect to various targets; predicting drug selectivity, toxicity profiles, ADMET and physicochemical properties; characteristics of drug-delivery systems and virtual screening. Neural networks continue to grow in importance for drug discovery. Recent developments in deep learning suggests further improvements may be gained in the analysis of large chemical data sets. It's anticipated that neural networks will be more widely used in drug discovery in the future, and applied in non-traditional areas such as drug delivery systems, biologically compatible materials, and regenerative medicine.

A Parallel Adaboost-Backpropagation Neural Network for Massive Image Dataset Classification

NASA Astrophysics Data System (ADS)

Cao, Jianfang; Chen, Lichao; Wang, Min; Shi, Hao; Tian, Yun

2016-12-01

Image classification uses computers to simulate human understanding and cognition of images by automatically categorizing images. This study proposes a faster image classification approach that parallelizes the traditional Adaboost-Backpropagation (BP) neural network using the MapReduce parallel programming model. First, we construct a strong classifier by assembling the outputs of 15 BP neural networks (which are individually regarded as weak classifiers) based on the Adaboost algorithm. Second, we design Map and Reduce tasks for both the parallel Adaboost-BP neural network and the feature extraction algorithm. Finally, we establish an automated classification model by building a Hadoop cluster. We use the Pascal VOC2007 and Caltech256 datasets to train and test the classification model. The results are superior to those obtained using traditional Adaboost-BP neural network or parallel BP neural network approaches. Our approach increased the average classification accuracy rate by approximately 14.5% and 26.0% compared to the traditional Adaboost-BP neural network and parallel BP neural network, respectively. Furthermore, the proposed approach requires less computation time and scales very well as evaluated by speedup, sizeup and scaleup. The proposed approach may provide a foundation for automated large-scale image classification and demonstrates practical value.

A Parallel Adaboost-Backpropagation Neural Network for Massive Image Dataset Classification.

PubMed

Cao, Jianfang; Chen, Lichao; Wang, Min; Shi, Hao; Tian, Yun

2016-12-01

Image classification uses computers to simulate human understanding and cognition of images by automatically categorizing images. This study proposes a faster image classification approach that parallelizes the traditional Adaboost-Backpropagation (BP) neural network using the MapReduce parallel programming model. First, we construct a strong classifier by assembling the outputs of 15 BP neural networks (which are individually regarded as weak classifiers) based on the Adaboost algorithm. Second, we design Map and Reduce tasks for both the parallel Adaboost-BP neural network and the feature extraction algorithm. Finally, we establish an automated classification model by building a Hadoop cluster. We use the Pascal VOC2007 and Caltech256 datasets to train and test the classification model. The results are superior to those obtained using traditional Adaboost-BP neural network or parallel BP neural network approaches. Our approach increased the average classification accuracy rate by approximately 14.5% and 26.0% compared to the traditional Adaboost-BP neural network and parallel BP neural network, respectively. Furthermore, the proposed approach requires less computation time and scales very well as evaluated by speedup, sizeup and scaleup. The proposed approach may provide a foundation for automated large-scale image classification and demonstrates practical value.

A Parallel Adaboost-Backpropagation Neural Network for Massive Image Dataset Classification

PubMed Central

Cao, Jianfang; Chen, Lichao; Wang, Min; Shi, Hao; Tian, Yun

2016-01-01

Image classification uses computers to simulate human understanding and cognition of images by automatically categorizing images. This study proposes a faster image classification approach that parallelizes the traditional Adaboost-Backpropagation (BP) neural network using the MapReduce parallel programming model. First, we construct a strong classifier by assembling the outputs of 15 BP neural networks (which are individually regarded as weak classifiers) based on the Adaboost algorithm. Second, we design Map and Reduce tasks for both the parallel Adaboost-BP neural network and the feature extraction algorithm. Finally, we establish an automated classification model by building a Hadoop cluster. We use the Pascal VOC2007 and Caltech256 datasets to train and test the classification model. The results are superior to those obtained using traditional Adaboost-BP neural network or parallel BP neural network approaches. Our approach increased the average classification accuracy rate by approximately 14.5% and 26.0% compared to the traditional Adaboost-BP neural network and parallel BP neural network, respectively. Furthermore, the proposed approach requires less computation time and scales very well as evaluated by speedup, sizeup and scaleup. The proposed approach may provide a foundation for automated large-scale image classification and demonstrates practical value. PMID:27905520

Deinterlacing using modular neural network

NASA Astrophysics Data System (ADS)

Woo, Dong H.; Eom, Il K.; Kim, Yoo S.

2004-05-01

Deinterlacing is the conversion process from the interlaced scan to progressive one. While many previous algorithms that are based on weighted-sum cause blurring in edge region, deinterlacing using neural network can reduce the blurring through recovering of high frequency component by learning process, and is found robust to noise. In proposed algorithm, input image is divided into edge and smooth region, and then, to each region, one neural network is assigned. Through this process, each neural network learns only patterns that are similar, therefore it makes learning more effective and estimation more accurate. But even within each region, there are various patterns such as long edge and texture in edge region. To solve this problem, modular neural network is proposed. In proposed modular neural network, two modules are combined in output node. One is for low frequency feature of local area of input image, and the other is for high frequency feature. With this structure, each modular neural network can learn different patterns with compensating for drawback of counterpart. Therefore it can adapt to various patterns within each region effectively. In simulation, the proposed algorithm shows better performance compared with conventional deinterlacing methods and single neural network method.

Pruning artificial neural networks using neural complexity measures.

PubMed

Jorgensen, Thomas D; Haynes, Barry P; Norlund, Charlotte C F

2008-10-01

This paper describes a new method for pruning artificial neural networks, using a measure of the neural complexity of the neural network. This measure is used to determine the connections that should be pruned. The measure computes the information-theoretic complexity of a neural network, which is similar to, yet different from previous research on pruning. The method proposed here shows how overly large and complex networks can be reduced in size, whilst retaining learnt behaviour and fitness. The technique proposed here helps to discover a network topology that matches the complexity of the problem it is meant to solve. This novel pruning technique is tested in a robot control domain, simulating a racecar. It is shown, that the proposed pruning method is a significant improvement over the most commonly used pruning method Magnitude Based Pruning. Furthermore, some of the pruned networks prove to be faster learners than the benchmark network that they originate from. This means that this pruning method can also help to unleash hidden potential in a network, because the learning time decreases substantially for a pruned a network, due to the reduction of dimensionality of the network.

Single-hidden-layer feed-forward quantum neural network based on Grover learning.

PubMed

Liu, Cheng-Yi; Chen, Chein; Chang, Ching-Ter; Shih, Lun-Min

2013-09-01

In this paper, a novel single-hidden-layer feed-forward quantum neural network model is proposed based on some concepts and principles in the quantum theory. By combining the quantum mechanism with the feed-forward neural network, we defined quantum hidden neurons and connected quantum weights, and used them as the fundamental information processing unit in a single-hidden-layer feed-forward neural network. The quantum neurons make a wide range of nonlinear functions serve as the activation functions in the hidden layer of the network, and the Grover searching algorithm outstands the optimal parameter setting iteratively and thus makes very efficient neural network learning possible. The quantum neuron and weights, along with a Grover searching algorithm based learning, result in a novel and efficient neural network characteristic of reduced network, high efficient training and prospect application in future. Some simulations are taken to investigate the performance of the proposed quantum network and the result show that it can achieve accurate learning. Copyright © 2013 Elsevier Ltd. All rights reserved.

Possible influences on clinical supervision.

PubMed

Jones, A

This article discusses clinical supervision and suggests that, aside from helping nurses to enhance their clinical effectiveness, it could offer experiential methods of assisting nurses to identify and locate supportive networks in the workplace. The advantages and difficulties of supervision relationships are described in context, including some consideration of authority and control.

Finite-time convergent recurrent neural network with a hard-limiting activation function for constrained optimization with piecewise-linear objective functions.

PubMed

Liu, Qingshan; Wang, Jun

2011-04-01

This paper presents a one-layer recurrent neural network for solving a class of constrained nonsmooth optimization problems with piecewise-linear objective functions. The proposed neural network is guaranteed to be globally convergent in finite time to the optimal solutions under a mild condition on a derived lower bound of a single gain parameter in the model. The number of neurons in the neural network is the same as the number of decision variables of the optimization problem. Compared with existing neural networks for optimization, the proposed neural network has a couple of salient features such as finite-time convergence and a low model complexity. Specific models for two important special cases, namely, linear programming and nonsmooth optimization, are also presented. In addition, applications to the shortest path problem and constrained least absolute deviation problem are discussed with simulation results to demonstrate the effectiveness and characteristics of the proposed neural network.

Financial Time Series Prediction Using Elman Recurrent Random Neural Networks

PubMed Central

Wang, Jie; Wang, Jun; Fang, Wen; Niu, Hongli

2016-01-01

In recent years, financial market dynamics forecasting has been a focus of economic research. To predict the price indices of stock markets, we developed an architecture which combined Elman recurrent neural networks with stochastic time effective function. By analyzing the proposed model with the linear regression, complexity invariant distance (CID), and multiscale CID (MCID) analysis methods and taking the model compared with different models such as the backpropagation neural network (BPNN), the stochastic time effective neural network (STNN), and the Elman recurrent neural network (ERNN), the empirical results show that the proposed neural network displays the best performance among these neural networks in financial time series forecasting. Further, the empirical research is performed in testing the predictive effects of SSE, TWSE, KOSPI, and Nikkei225 with the established model, and the corresponding statistical comparisons of the above market indices are also exhibited. The experimental results show that this approach gives good performance in predicting the values from the stock market indices. PMID:27293423

Periodicity and stability for variable-time impulsive neural networks.

PubMed

Li, Hongfei; Li, Chuandong; Huang, Tingwen

2017-10-01

The paper considers a general neural networks model with variable-time impulses. It is shown that each solution of the system intersects with every discontinuous surface exactly once via several new well-proposed assumptions. Moreover, based on the comparison principle, this paper shows that neural networks with variable-time impulse can be reduced to the corresponding neural network with fixed-time impulses under well-selected conditions. Meanwhile, the fixed-time impulsive systems can be regarded as the comparison system of the variable-time impulsive neural networks. Furthermore, a series of sufficient criteria are derived to ensure the existence and global exponential stability of periodic solution of variable-time impulsive neural networks, and to illustrate the same stability properties between variable-time impulsive neural networks and the fixed-time ones. The new criteria are established by applying Schaefer's fixed point theorem combined with the use of inequality technique. Finally, a numerical example is presented to show the effectiveness of the proposed results. Copyright © 2017 Elsevier Ltd. All rights reserved.

Linear and nonlinear ARMA model parameter estimation using an artificial neural network

NASA Technical Reports Server (NTRS)

Chon, K. H.; Cohen, R. J.

1997-01-01

This paper addresses parametric system identification of linear and nonlinear dynamic systems by analysis of the input and output signals. Specifically, we investigate the relationship between estimation of the system using a feedforward neural network model and estimation of the system by use of linear and nonlinear autoregressive moving-average (ARMA) models. By utilizing a neural network model incorporating a polynomial activation function, we show the equivalence of the artificial neural network to the linear and nonlinear ARMA models. We compare the parameterization of the estimated system using the neural network and ARMA approaches by utilizing data generated by means of computer simulations. Specifically, we show that the parameters of a simulated ARMA system can be obtained from the neural network analysis of the simulated data or by conventional least squares ARMA analysis. The feasibility of applying neural networks with polynomial activation functions to the analysis of experimental data is explored by application to measurements of heart rate (HR) and instantaneous lung volume (ILV) fluctuations.

Financial Time Series Prediction Using Elman Recurrent Random Neural Networks.

PubMed

Wang, Jie; Wang, Jun; Fang, Wen; Niu, Hongli

2016-01-01

In recent years, financial market dynamics forecasting has been a focus of economic research. To predict the price indices of stock markets, we developed an architecture which combined Elman recurrent neural networks with stochastic time effective function. By analyzing the proposed model with the linear regression, complexity invariant distance (CID), and multiscale CID (MCID) analysis methods and taking the model compared with different models such as the backpropagation neural network (BPNN), the stochastic time effective neural network (STNN), and the Elman recurrent neural network (ERNN), the empirical results show that the proposed neural network displays the best performance among these neural networks in financial time series forecasting. Further, the empirical research is performed in testing the predictive effects of SSE, TWSE, KOSPI, and Nikkei225 with the established model, and the corresponding statistical comparisons of the above market indices are also exhibited. The experimental results show that this approach gives good performance in predicting the values from the stock market indices.

A novel neural-wavelet approach for process diagnostics and complex system modeling

NASA Astrophysics Data System (ADS)

Gao, Rong

Neural networks have been effective in several engineering applications because of their learning abilities and robustness. However certain shortcomings, such as slow convergence and local minima, are always associated with neural networks, especially neural networks applied to highly nonlinear and non-stationary problems. These problems can be effectively alleviated by integrating a new powerful tool, wavelets, into conventional neural networks. The multi-resolution analysis and feature localization capabilities of the wavelet transform offer neural networks new possibilities for learning. A neural wavelet network approach developed in this thesis enjoys fast convergence rate with little possibility to be caught at a local minimum. It combines the localization properties of wavelets with the learning abilities of neural networks. Two different testbeds are used for testing the efficiency of the new approach. The first is magnetic flowmeter-based process diagnostics: here we extend previous work, which has demonstrated that wavelet groups contain process information, to more general process diagnostics. A loop at Applied Intelligent Systems Lab (AISL) is used for collecting and analyzing data through the neural-wavelet approach. The research is important for thermal-hydraulic processes in nuclear and other engineering fields. The neural-wavelet approach developed is also tested with data from the electric power grid. More specifically, the neural-wavelet approach is used for performing short-term and mid-term prediction of power load demand. In addition, the feasibility of determining the type of load using the proposed neural wavelet approach is also examined. The notion of cross scale product has been developed as an expedient yet reliable discriminator of loads. Theoretical issues involved in the integration of wavelets and neural networks are discussed and future work outlined.

Active Control of Wind-Tunnel Model Aeroelastic Response Using Neural Networks

NASA Technical Reports Server (NTRS)

Scott, Robert C.

2000-01-01

NASA Langley Research Center, Hampton, VA 23681 Under a joint research and development effort conducted by the National Aeronautics and Space Administration and The Boeing Company (formerly McDonnell Douglas) three neural-network based control systems were developed and tested. The control systems were experimentally evaluated using a transonic wind-tunnel model in the Langley Transonic Dynamics Tunnel. One system used a neural network to schedule flutter suppression control laws, another employed a neural network in a predictive control scheme, and the third employed a neural network in an inverse model control scheme. All three of these control schemes successfully suppressed flutter to or near the limits of the testing apparatus, and represent the first experimental applications of neural networks to flutter suppression. This paper will summarize the findings of this project.

Modeling Aircraft Wing Loads from Flight Data Using Neural Networks

NASA Technical Reports Server (NTRS)

Allen, Michael J.; Dibley, Ryan P.

2003-01-01

Neural networks were used to model wing bending-moment loads, torsion loads, and control surface hinge-moments of the Active Aeroelastic Wing (AAW) aircraft. Accurate loads models are required for the development of control laws designed to increase roll performance through wing twist while not exceeding load limits. Inputs to the model include aircraft rates, accelerations, and control surface positions. Neural networks were chosen to model aircraft loads because they can account for uncharacterized nonlinear effects while retaining the capability to generalize. The accuracy of the neural network models was improved by first developing linear loads models to use as starting points for network training. Neural networks were then trained with flight data for rolls, loaded reversals, wind-up-turns, and individual control surface doublets for load excitation. Generalization was improved by using gain weighting and early stopping. Results are presented for neural network loads models of four wing loads and four control surface hinge moments at Mach 0.90 and an altitude of 15,000 ft. An average model prediction error reduction of 18.6 percent was calculated for the neural network models when compared to the linear models. This paper documents the input data conditioning, input parameter selection, structure, training, and validation of the neural network models.

Imaging spectroscopy in soil-water based site suitability assessment for artificial regeneration to Scots pine

NASA Astrophysics Data System (ADS)

Middleton, Maarit; Närhi, Paavo; Sutinen, Raimo

In a humid northern boreal climate, the success rate of artificial regeneration to Scots pine ( Pinus sylvestris L.) can be improved by including a soil water content (SWC) based assessment of site suitability in the reforestation planning process. This paper introduces an application of airborne visible-near-infrared imaging spectroscopic data to identify suitable subregions of forest compartments for the low SWC-tolerant Scots pine. The spatial patterns of understorey plant species communities, recorded by the AISA (Airborne Imaging Spectrometer for Applications) sensor, were demonstrated to be dependant on the underlying SWC. According to the nonmetric multidimensional scaling and correlation results twelve understorey species were found to be most abundant on sites with high soil SWCs. The abundance of bare soil, rocks and abundance of more than ten species indicated low soil SWCs. The spatial patterns of understorey are attributed to time-stability of the underlying SWC patterns. A supervised artificial neural network (radial basis functional link network, probabilistic neural network) approach was taken to classify AISA imaging spectrometer data with dielectric (as a measure volumetric SWC) ground referencing into regimes suitable and unsuitable for Scots pine. The accuracy assessment with receiver operating characteristics curves demonstrated a maximum of 74.1% area under the curve values which indicated moderate success of the NN modelling. The results signified the importance of the training set's quality, adequate quantity (>2.43 points/ha) and NN algorithm selection over the NN algorithm training parameter optimization to perfection. This methodology for the analysis of site suitability of Scots pine can be recommended, especially when artificial regeneration of former mixed wood Norway spruce ( Picea abies L. Karst) - downy birch ( Betula pubenscens Ehrh.) stands is being considered, so that artificially regenerated areas to Scots pine can be optimized for forestry purposes.

A Neuromorphic Architecture for Object Recognition and Motion Anticipation Using Burst-STDP

PubMed Central

Balduzzi, David; Tononi, Giulio

2012-01-01

In this work we investigate the possibilities offered by a minimal framework of artificial spiking neurons to be deployed in silico. Here we introduce a hierarchical network architecture of spiking neurons which learns to recognize moving objects in a visual environment and determine the correct motor output for each object. These tasks are learned through both supervised and unsupervised spike timing dependent plasticity (STDP). STDP is responsible for the strengthening (or weakening) of synapses in relation to pre- and post-synaptic spike times and has been described as a Hebbian paradigm taking place both in vitro and in vivo. We utilize a variation of STDP learning, called burst-STDP, which is based on the notion that, since spikes are expensive in terms of energy consumption, then strong bursting activity carries more information than single (sparse) spikes. Furthermore, this learning algorithm takes advantage of homeostatic renormalization, which has been hypothesized to promote memory consolidation during NREM sleep. Using this learning rule, we design a spiking neural network architecture capable of object recognition, motion detection, attention towards important objects, and motor control outputs. We demonstrate the abilities of our design in a simple environment with distractor objects, multiple objects moving concurrently, and in the presence of noise. Most importantly, we show how this neural network is capable of performing these tasks using a simple leaky-integrate-and-fire (LIF) neuron model with binary synapses, making it fully compatible with state-of-the-art digital neuromorphic hardware designs. As such, the building blocks and learning rules presented in this paper appear promising for scalable fully neuromorphic systems to be implemented in hardware chips. PMID:22615855

Exponential H(infinity) synchronization of general discrete-time chaotic neural networks with or without time delays.

PubMed

Qi, Donglian; Liu, Meiqin; Qiu, Meikang; Zhang, Senlin

2010-08-01

This brief studies exponential H(infinity) synchronization of a class of general discrete-time chaotic neural networks with external disturbance. On the basis of the drive-response concept and H(infinity) control theory, and using Lyapunov-Krasovskii (or Lyapunov) functional, state feedback controllers are established to not only guarantee exponential stable synchronization between two general chaotic neural networks with or without time delays, but also reduce the effect of external disturbance on the synchronization error to a minimal H(infinity) norm constraint. The proposed controllers can be obtained by solving the convex optimization problems represented by linear matrix inequalities. Most discrete-time chaotic systems with or without time delays, such as Hopfield neural networks, cellular neural networks, bidirectional associative memory networks, recurrent multilayer perceptrons, Cohen-Grossberg neural networks, Chua's circuits, etc., can be transformed into this general chaotic neural network to be H(infinity) synchronization controller designed in a unified way. Finally, some illustrated examples with their simulations have been utilized to demonstrate the effectiveness of the proposed methods.

Automated implementation of rule-based expert systems with neural networks for time-critical applications

NASA Technical Reports Server (NTRS)

Ramamoorthy, P. A.; Huang, Song; Govind, Girish

1991-01-01

In fault diagnosis, control and real-time monitoring, both timing and accuracy are critical for operators or machines to reach proper solutions or appropriate actions. Expert systems are becoming more popular in the manufacturing community for dealing with such problems. In recent years, neural networks have revived and their applications have spread to many areas of science and engineering. A method of using neural networks to implement rule-based expert systems for time-critical applications is discussed here. This method can convert a given rule-based system into a neural network with fixed weights and thresholds. The rules governing the translation are presented along with some examples. We also present the results of automated machine implementation of such networks from the given rule-base. This significantly simplifies the translation process to neural network expert systems from conventional rule-based systems. Results comparing the performance of the proposed approach based on neural networks vs. the classical approach are given. The possibility of very large scale integration (VLSI) realization of such neural network expert systems is also discussed.

Predicting Slag Generation in Sub-Scale Test Motors Using a Neural Network

NASA Technical Reports Server (NTRS)

Wiesenberg, Brent

1999-01-01

Generation of slag (aluminum oxide) is an important issue for the Reusable Solid Rocket Motor (RSRM). Thiokol performed testing to quantify the relationship between raw material variations and slag generation in solid propellants by testing sub-scale motors cast with propellant containing various combinations of aluminum fuel and ammonium perchlorate (AP) oxidizer particle sizes. The test data were analyzed using statistical methods and an artificial neural network. This paper primarily addresses the neural network results with some comparisons to the statistical results. The neural network showed that the particle sizes of both the aluminum and unground AP have a measurable effect on slag generation. The neural network analysis showed that aluminum particle size is the dominant driver in slag generation, about 40% more influential than AP. The network predictions of the amount of slag produced during firing of sub-scale motors were 16% better than the predictions of a statistically derived empirical equation. Another neural network successfully characterized the slag generated during full-scale motor tests. The success is attributable to the ability of neural networks to characterize multiple complex factors including interactions that affect slag generation.

Application of Two-Dimensional AWE Algorithm in Training Multi-Dimensional Neural Network Model

DTIC Science & Technology

2003-07-01

hybrid scheme . the general neural network method (Table 3.1). The training process of the software- ACKNOWLEDGMENT "Neuralmodeler" is shown in Fig. 3.2...engineering. Artificial neural networks (ANNs) have emerged Training a neural network model is the key of as a powerful technique for modeling general neural...coefficients am, the derivatives method of moments (MoM). The variables in the of matrix I have to be generated . A closed form model are frequency

Center for Neural Engineering at Tennessee State University, ASSERT Annual Progress Report.

DTIC Science & Technology

1995-07-01

neural networks . Their research topics are: (1) developing frequency dependent oscillatory neural networks ; (2) long term pontentiation learning rules...as applied to spatial navigation; (3) design and build a servo joint robotic arm and (4) neural network based prothesis control. One graduate student

A Feasibility Study of Synthesizing Subsurfaces Modeled with Computational Neural Networks

NASA Technical Reports Server (NTRS)

Wang, John T.; Housner, Jerrold M.; Szewczyk, Z. Peter

1998-01-01

This paper investigates the feasibility of synthesizing substructures modeled with computational neural networks. Substructures are modeled individually with computational neural networks and the response of the assembled structure is predicted by synthesizing the neural networks. A superposition approach is applied to synthesize models for statically determinate substructures while an interface displacement collocation approach is used to synthesize statically indeterminate substructure models. Beam and plate substructures along with components of a complicated Next Generation Space Telescope (NGST) model are used in this feasibility study. In this paper, the limitations and difficulties of synthesizing substructures modeled with neural networks are also discussed.

Optical-Correlator Neural Network Based On Neocognitron

NASA Technical Reports Server (NTRS)

Chao, Tien-Hsin; Stoner, William W.

1994-01-01

Multichannel optical correlator implements shift-invariant, high-discrimination pattern-recognizing neural network based on paradigm of neocognitron. Selected as basic building block of this neural network because invariance under shifts is inherent advantage of Fourier optics included in optical correlators in general. Neocognitron is conceptual electronic neural-network model for recognition of visual patterns. Multilayer processing achieved by iteratively feeding back output of feature correlator to input spatial light modulator and updating Fourier filters. Neural network trained by use of characteristic features extracted from target images. Multichannel implementation enables parallel processing of large number of selected features.

Neural network based system for equipment surveillance

DOEpatents

Vilim, Richard B.; Gross, Kenneth C.; Wegerich, Stephan W.

1998-01-01

A method and system for performing surveillance of transient signals of an industrial device to ascertain the operating state. The method and system involves the steps of reading into a memory training data, determining neural network weighting values until achieving target outputs close to the neural network output. If the target outputs are inadequate, wavelet parameters are determined to yield neural network outputs close to the desired set of target outputs and then providing signals characteristic of an industrial process and comparing the neural network output to the industrial process signals to evaluate the operating state of the industrial process.

Neural network based system for equipment surveillance

DOEpatents

Vilim, R.B.; Gross, K.C.; Wegerich, S.W.

1998-04-28

A method and system are disclosed for performing surveillance of transient signals of an industrial device to ascertain the operating state. The method and system involves the steps of reading into a memory training data, determining neural network weighting values until achieving target outputs close to the neural network output. If the target outputs are inadequate, wavelet parameters are determined to yield neural network outputs close to the desired set of target outputs and then providing signals characteristic of an industrial process and comparing the neural network output to the industrial process signals to evaluate the operating state of the industrial process. 33 figs.

Neural networks for function approximation in nonlinear control

NASA Technical Reports Server (NTRS)

Linse, Dennis J.; Stengel, Robert F.

1990-01-01

Two neural network architectures are compared with a classical spline interpolation technique for the approximation of functions useful in a nonlinear control system. A standard back-propagation feedforward neural network and a cerebellar model articulation controller (CMAC) neural network are presented, and their results are compared with a B-spline interpolation procedure that is updated using recursive least-squares parameter identification. Each method is able to accurately represent a one-dimensional test function. Tradeoffs between size requirements, speed of operation, and speed of learning indicate that neural networks may be practical for identification and adaptation in a nonlinear control environment.

Vibrational Analysis of Engine Components Using Neural-Net Processing and Electronic Holography

NASA Technical Reports Server (NTRS)

Decker, Arthur J.; Fite, E. Brian; Mehmed, Oral; Thorp, Scott A.

1997-01-01

The use of computational-model trained artificial neural networks to acquire damage specific information from electronic holograms is discussed. A neural network is trained to transform two time-average holograms into a pattern related to the bending-induced-strain distribution of the vibrating component. The bending distribution is very sensitive to component damage unlike the characteristic fringe pattern or the displacement amplitude distribution. The neural network processor is fast for real-time visualization of damage. The two-hologram limit makes the processor more robust to speckle pattern decorrelation. Undamaged and cracked cantilever plates serve as effective objects for testing the combination of electronic holography and neural-net processing. The requirements are discussed for using finite-element-model trained neural networks for field inspections of engine components. The paper specifically discusses neural-network fringe pattern analysis in the presence of the laser speckle effect and the performances of two limiting cases of the neural-net architecture.

Vibrational Analysis of Engine Components Using Neural-Net Processing and Electronic Holography

NASA Technical Reports Server (NTRS)

Decker, Arthur J.; Fite, E. Brian; Mehmed, Oral; Thorp, Scott A.

1998-01-01

The use of computational-model trained artificial neural networks to acquire damage specific information from electronic holograms is discussed. A neural network is trained to transform two time-average holograms into a pattern related to the bending-induced-strain distribution of the vibrating component. The bending distribution is very sensitive to component damage unlike the characteristic fringe pattern or the displacement amplitude distribution. The neural network processor is fast for real-time visualization of damage. The two-hologram limit makes the processor more robust to speckle pattern decorrelation. Undamaged and cracked cantilever plates serve as effective objects for testing the combination of electronic holography and neural-net processing. The requirements are discussed for using finite-element-model trained neural networks for field inspections of engine components. The paper specifically discusses neural-network fringe pattern analysis in the presence of the laser speckle effect and the performances of two limiting cases of the neural-net architecture.

Deep convolutional neural network for mammographic density segmentation

NASA Astrophysics Data System (ADS)

Wei, Jun; Li, Songfeng; Chan, Heang-Ping; Helvie, Mark A.; Roubidoux, Marilyn A.; Lu, Yao; Zhou, Chuan; Hadjiiski, Lubomir; Samala, Ravi K.

2018-02-01

Breast density is one of the most significant factors for cancer risk. In this study, we proposed a supervised deep learning approach for automated estimation of percentage density (PD) on digital mammography (DM). The deep convolutional neural network (DCNN) was trained to estimate a probability map of breast density (PMD). PD was calculated as the ratio of the dense area to the breast area based on the probability of each pixel belonging to dense region or fatty region at a decision threshold of 0.5. The DCNN estimate was compared to a feature-based statistical learning approach, in which gray level, texture and morphological features were extracted from each ROI and the least absolute shrinkage and selection operator (LASSO) was used to select and combine the useful features to generate the PMD. The reference PD of each image was provided by two experienced MQSA radiologists. With IRB approval, we retrospectively collected 347 DMs from patient files at our institution. The 10-fold cross-validation results showed a strong correlation r=0.96 between the DCNN estimation and interactive segmentation by radiologists while that of the feature-based statistical learning approach vs radiologists' segmentation had a correlation r=0.78. The difference between the segmentation by DCNN and by radiologists was significantly smaller than that between the feature-based learning approach and radiologists (p < 0.0001) by two-tailed paired t-test. This study demonstrated that the DCNN approach has the potential to replace radiologists' interactive thresholding in PD estimation on DMs.

Multi-task transfer learning deep convolutional neural network: application to computer-aided diagnosis of breast cancer on mammograms

NASA Astrophysics Data System (ADS)

Samala, Ravi K.; Chan, Heang-Ping; Hadjiiski, Lubomir M.; Helvie, Mark A.; Cha, Kenny H.; Richter, Caleb D.

2017-12-01

Transfer learning in deep convolutional neural networks (DCNNs) is an important step in its application to medical imaging tasks. We propose a multi-task transfer learning DCNN with the aim of translating the ‘knowledge’ learned from non-medical images to medical diagnostic tasks through supervised training and increasing the generalization capabilities of DCNNs by simultaneously learning auxiliary tasks. We studied this approach in an important application: classification of malignant and benign breast masses. With Institutional Review Board (IRB) approval, digitized screen-film mammograms (SFMs) and digital mammograms (DMs) were collected from our patient files and additional SFMs were obtained from the Digital Database for Screening Mammography. The data set consisted of 2242 views with 2454 masses (1057 malignant, 1397 benign). In single-task transfer learning, the DCNN was trained and tested on SFMs. In multi-task transfer learning, SFMs and DMs were used to train the DCNN, which was then tested on SFMs. N-fold cross-validation with the training set was used for training and parameter optimization. On the independent test set, the multi-task transfer learning DCNN was found to have significantly (p  =  0.007) higher performance compared to the single-task transfer learning DCNN. This study demonstrates that multi-task transfer learning may be an effective approach for training DCNN in medical imaging applications when training samples from a single modality are limited.

Deep Learning of Atomically Resolved Scanning Transmission Electron Microscopy Images: Chemical Identification and Tracking Local Transformations.

PubMed

Ziatdinov, Maxim; Dyck, Ondrej; Maksov, Artem; Li, Xufan; Sang, Xiahan; Xiao, Kai; Unocic, Raymond R; Vasudevan, Rama; Jesse, Stephen; Kalinin, Sergei V

2017-12-26

Recent advances in scanning transmission electron and scanning probe microscopies have opened exciting opportunities in probing the materials structural parameters and various functional properties in real space with angstrom-level precision. This progress has been accompanied by an exponential increase in the size and quality of data sets produced by microscopic and spectroscopic experimental techniques. These developments necessitate adequate methods for extracting relevant physical and chemical information from the large data sets, for which a priori information on the structures of various atomic configurations and lattice defects is limited or absent. Here we demonstrate an application of deep neural networks to extract information from atomically resolved images including location of the atomic species and type of defects. We develop a "weakly supervised" approach that uses information on the coordinates of all atomic species in the image, extracted via a deep neural network, to identify a rich variety of defects that are not part of an initial training set. We further apply our approach to interpret complex atomic and defect transformation, including switching between different coordination of silicon dopants in graphene as a function of time, formation of peculiar silicon dimer with mixed 3-fold and 4-fold coordination, and the motion of molecular "rotor". This deep learning-based approach resembles logic of a human operator, but can be scaled leading to significant shift in the way of extracting and analyzing information from raw experimental data.

Non-metallic coating thickness prediction using artificial neural network and support vector machine with time resolved thermography

NASA Astrophysics Data System (ADS)

Wang, Hongjin; Hsieh, Sheng-Jen; Peng, Bo; Zhou, Xunfei

2016-07-01

A method without requirements on knowledge about thermal properties of coatings or those of substrates will be interested in the industrial application. Supervised machine learning regressions may provide possible solution to the problem. This paper compares the performances of two regression models (artificial neural networks (ANN) and support vector machines for regression (SVM)) with respect to coating thickness estimations made based on surface temperature increments collected via time resolved thermography. We describe SVM roles in coating thickness prediction. Non-dimensional analyses are conducted to illustrate the effects of coating thicknesses and various factors on surface temperature increments. It's theoretically possible to correlate coating thickness with surface increment. Based on the analyses, the laser power is selected in such a way: during the heating, the temperature increment is high enough to determine the coating thickness variance but low enough to avoid surface melting. Sixty-one pain-coated samples with coating thicknesses varying from 63.5 μm to 571 μm are used to train models. Hyper-parameters of the models are optimized by 10-folder cross validation. Another 28 sets of data are then collected to test the performance of the three methods. The study shows that SVM can provide reliable predictions of unknown data, due to its deterministic characteristics, and it works well when used for a small input data group. The SVM model generates more accurate coating thickness estimates than the ANN model.

Unsupervised Learning —A Novel Clustering Method for Rolling Bearing Faults Identification

NASA Astrophysics Data System (ADS)

Kai, Li; Bo, Luo; Tao, Ma; Xuefeng, Yang; Guangming, Wang

2017-12-01

To promptly process the massive fault data and automatically provide accurate diagnosis results, numerous studies have been conducted on intelligent fault diagnosis of rolling bearing. Among these studies, such as artificial neural networks, support vector machines, decision trees and other supervised learning methods are used commonly. These methods can detect the failure of rolling bearing effectively, but to achieve better detection results, it often requires a lot of training samples. Based on above, a novel clustering method is proposed in this paper. This novel method is able to find the correct number of clusters automatically the effectiveness of the proposed method is validated using datasets from rolling element bearings. The diagnosis results show that the proposed method can accurately detect the fault types of small samples. Meanwhile, the diagnosis results are also relative high accuracy even for massive samples.

Application of the Actor-Critic Architecture to Functional Electrical Stimulation Control of a Human Arm

PubMed Central

Thomas, Philip; Branicky, Michael; van den Bogert, Antonie; Jagodnik, Kathleen

2010-01-01

Clinical tests have shown that the dynamics of a human arm, controlled using Functional Electrical Stimulation (FES), can vary significantly between and during trials. In this paper, we study the application of the actor-critic architecture, with neural networks for the both the actor and the critic, as a controller that can adapt to these changing dynamics of a human arm. Development and tests were done in simulation using a planar arm model and Hill-based muscle dynamics. We begin by training it using a Proportional Derivative (PD) controller as a supervisor. We then make clinically relevant changes to the dynamics of the arm and test the actor-critic’s ability to adapt without supervision in a reasonable number of episodes. Finally, we devise methods for achieving both rapid learning and long-term stability. PMID:20689654

Application of the Actor-Critic Architecture to Functional Electrical Stimulation Control of a Human Arm.

PubMed

Thomas, Philip; Branicky, Michael; van den Bogert, Antonie; Jagodnik, Kathleen

2009-01-01

Clinical tests have shown that the dynamics of a human arm, controlled using Functional Electrical Stimulation (FES), can vary significantly between and during trials. In this paper, we study the application of the actor-critic architecture, with neural networks for the both the actor and the critic, as a controller that can adapt to these changing dynamics of a human arm. Development and tests were done in simulation using a planar arm model and Hill-based muscle dynamics. We begin by training it using a Proportional Derivative (PD) controller as a supervisor. We then make clinically relevant changes to the dynamics of the arm and test the actor-critic's ability to adapt without supervision in a reasonable number of episodes. Finally, we devise methods for achieving both rapid learning and long-term stability.

Learning the Relationship between Galaxy Spectra and Star Formation Histories

NASA Astrophysics Data System (ADS)

Lovell, Christopher; Acquaviva, Viviana; Iyer, Kartheik; Gawiser, Eric

2018-01-01

We explore novel approaches to the problem of predicting a galaxy’s star formation history (SFH) from its Spectral Energy Distribution (SED). Traditional approaches to SED template fitting use constant or exponentially declining SFHs, and are known to incur significant bias in the inferred SFHs, which are typically skewed toward younger stellar populations. Machine learning approaches, including tree ensemble methods and convolutional neural networks, would not be affected by the same bias, and may work well in recovering unbiased and multi-episodic star formation histories. We use a supervised approach whereby models are trained using synthetic spectra, generated from three state of the art hydrodynamical simulations, including nebular emission. We explore how SED feature maps can be used to highlight areas of the spectrum with the highest predictive power and discuss the limitations of the approach when applied to real data.

Neural networks for vertical microcode compaction

NASA Astrophysics Data System (ADS)

Chu, Pong P.

1992-09-01

Neural networks provide an alternative way to solve complex optimization problems. Instead of performing a program of instructions sequentially as in a traditional computer, neural network model explores many competing hypotheses simultaneously using its massively parallel net. The paper shows how to use the neural network approach to perform vertical micro-code compaction for a micro-programmed control unit. The compaction procedure includes two basic steps. The first step determines the compatibility classes and the second step selects a minimal subset to cover the control signals. Since the selection process is an NP- complete problem, to find an optimal solution is impractical. In this study, we employ a customized neural network to obtain the minimal subset. We first formalize this problem, and then define an `energy function' and map it to a two-layer fully connected neural network. The modified network has two types of neurons and can always obtain a valid solution.

Advances in Artificial Neural Networks - Methodological Development and Application

USDA-ARS?s Scientific Manuscript database

Artificial neural networks as a major soft-computing technology have been extensively studied and applied during the last three decades. Research on backpropagation training algorithms for multilayer perceptron networks has spurred development of other neural network training algorithms for other ne...

Semi-supervised spectral algorithms for community detection in complex networks based on equivalence of clustering methods

NASA Astrophysics Data System (ADS)

Ma, Xiaoke; Wang, Bingbo; Yu, Liang

2018-01-01

Community detection is fundamental for revealing the structure-functionality relationship in complex networks, which involves two issues-the quantitative function for community as well as algorithms to discover communities. Despite significant research on either of them, few attempt has been made to establish the connection between the two issues. To attack this problem, a generalized quantification function is proposed for community in weighted networks, which provides a framework that unifies several well-known measures. Then, we prove that the trace optimization of the proposed measure is equivalent with the objective functions of algorithms such as nonnegative matrix factorization, kernel K-means as well as spectral clustering. It serves as the theoretical foundation for designing algorithms for community detection. On the second issue, a semi-supervised spectral clustering algorithm is developed by exploring the equivalence relation via combining the nonnegative matrix factorization and spectral clustering. Different from the traditional semi-supervised algorithms, the partial supervision is integrated into the objective of the spectral algorithm. Finally, through extensive experiments on both artificial and real world networks, we demonstrate that the proposed method improves the accuracy of the traditional spectral algorithms in community detection.

Artificial Neural Network Metamodels of Stochastic Computer Simulations

DTIC Science & Technology

1994-08-10

SUBTITLE r 5. FUNDING NUMBERS Artificial Neural Network Metamodels of Stochastic I () Computer Simulations 6. AUTHOR(S) AD- A285 951 Robert Allen...8217!298*1C2 ARTIFICIAL NEURAL NETWORK METAMODELS OF STOCHASTIC COMPUTER SIMULATIONS by Robert Allen Kilmer B.S. in Education Mathematics, Indiana...dedicate this document to the memory of my father, William Ralph Kilmer. mi ABSTRACT Signature ARTIFICIAL NEURAL NETWORK METAMODELS OF STOCHASTIC

Research on wind field algorithm of wind lidar based on BP neural network and grey prediction

NASA Astrophysics Data System (ADS)

Chen, Yong; Chen, Chun-Li; Luo, Xiong; Zhang, Yan; Yang, Ze-hou; Zhou, Jie; Shi, Xiao-ding; Wang, Lei

2018-01-01

This paper uses the BP neural network and grey algorithm to forecast and study radar wind field. In order to reduce the residual error in the wind field prediction which uses BP neural network and grey algorithm, calculating the minimum value of residual error function, adopting the residuals of the gray algorithm trained by BP neural network, using the trained network model to forecast the residual sequence, using the predicted residual error sequence to modify the forecast sequence of the grey algorithm. The test data show that using the grey algorithm modified by BP neural network can effectively reduce the residual value and improve the prediction precision.

An Intelligent Ensemble Neural Network Model for Wind Speed Prediction in Renewable Energy Systems.

PubMed

Ranganayaki, V; Deepa, S N

2016-01-01

Various criteria are proposed to select the number of hidden neurons in artificial neural network (ANN) models and based on the criterion evolved an intelligent ensemble neural network model is proposed to predict wind speed in renewable energy applications. The intelligent ensemble neural model based wind speed forecasting is designed by averaging the forecasted values from multiple neural network models which includes multilayer perceptron (MLP), multilayer adaptive linear neuron (Madaline), back propagation neural network (BPN), and probabilistic neural network (PNN) so as to obtain better accuracy in wind speed prediction with minimum error. The random selection of hidden neurons numbers in artificial neural network results in overfitting or underfitting problem. This paper aims to avoid the occurrence of overfitting and underfitting problems. The selection of number of hidden neurons is done in this paper employing 102 criteria; these evolved criteria are verified by the computed various error values. The proposed criteria for fixing hidden neurons are validated employing the convergence theorem. The proposed intelligent ensemble neural model is applied for wind speed prediction application considering the real time wind data collected from the nearby locations. The obtained simulation results substantiate that the proposed ensemble model reduces the error value to minimum and enhances the accuracy. The computed results prove the effectiveness of the proposed ensemble neural network (ENN) model with respect to the considered error factors in comparison with that of the earlier models available in the literature.

An Intelligent Ensemble Neural Network Model for Wind Speed Prediction in Renewable Energy Systems

PubMed Central

Ranganayaki, V.; Deepa, S. N.

2016-01-01

Various criteria are proposed to select the number of hidden neurons in artificial neural network (ANN) models and based on the criterion evolved an intelligent ensemble neural network model is proposed to predict wind speed in renewable energy applications. The intelligent ensemble neural model based wind speed forecasting is designed by averaging the forecasted values from multiple neural network models which includes multilayer perceptron (MLP), multilayer adaptive linear neuron (Madaline), back propagation neural network (BPN), and probabilistic neural network (PNN) so as to obtain better accuracy in wind speed prediction with minimum error. The random selection of hidden neurons numbers in artificial neural network results in overfitting or underfitting problem. This paper aims to avoid the occurrence of overfitting and underfitting problems. The selection of number of hidden neurons is done in this paper employing 102 criteria; these evolved criteria are verified by the computed various error values. The proposed criteria for fixing hidden neurons are validated employing the convergence theorem. The proposed intelligent ensemble neural model is applied for wind speed prediction application considering the real time wind data collected from the nearby locations. The obtained simulation results substantiate that the proposed ensemble model reduces the error value to minimum and enhances the accuracy. The computed results prove the effectiveness of the proposed ensemble neural network (ENN) model with respect to the considered error factors in comparison with that of the earlier models available in the literature. PMID:27034973

SPECIAL ISSUE ON OPTICAL PROCESSING OF INFORMATION: Optical neural networks based on holographic correlators

NASA Astrophysics Data System (ADS)

Sokolov, V. K.; Shubnikov, E. I.

1995-10-01

The three most important models of neural networks — a bidirectional associative memory, Hopfield networks, and adaptive resonance networks — are used as examples to show that a holographic correlator has its place in the neural computing paradigm.

Comparison of artificial intelligence classifiers for SIP attack data

NASA Astrophysics Data System (ADS)

Safarik, Jakub; Slachta, Jiri

2016-05-01

Honeypot application is a source of valuable data about attacks on the network. We run several SIP honeypots in various computer networks, which are separated geographically and logically. Each honeypot runs on public IP address and uses standard SIP PBX ports. All information gathered via honeypot is periodically sent to the centralized server. This server classifies all attack data by neural network algorithm. The paper describes optimizations of a neural network classifier, which lower the classification error. The article contains the comparison of two neural network algorithm used for the classification of validation data. The first is the original implementation of the neural network described in recent work; the second neural network uses further optimizations like input normalization or cross-entropy cost function. We also use other implementations of neural networks and machine learning classification algorithms. The comparison test their capabilities on validation data to find the optimal classifier. The article result shows promise for further development of an accurate SIP attack classification engine.

Parallel consensual neural networks.

PubMed

Benediktsson, J A; Sveinsson, J R; Ersoy, O K; Swain, P H

1997-01-01

A new type of a neural-network architecture, the parallel consensual neural network (PCNN), is introduced and applied in classification/data fusion of multisource remote sensing and geographic data. The PCNN architecture is based on statistical consensus theory and involves using stage neural networks with transformed input data. The input data are transformed several times and the different transformed data are used as if they were independent inputs. The independent inputs are first classified using the stage neural networks. The output responses from the stage networks are then weighted and combined to make a consensual decision. In this paper, optimization methods are used in order to weight the outputs from the stage networks. Two approaches are proposed to compute the data transforms for the PCNN, one for binary data and another for analog data. The analog approach uses wavelet packets. The experimental results obtained with the proposed approach show that the PCNN outperforms both a conjugate-gradient backpropagation neural network and conventional statistical methods in terms of overall classification accuracy of test data.

Unfolding the neutron spectrum of a NE213 scintillator using artificial neural networks.

PubMed

Sharghi Ido, A; Bonyadi, M R; Etaati, G R; Shahriari, M

2009-10-01

Artificial neural networks technology has been applied to unfold the neutron spectra from the pulse height distribution measured with NE213 liquid scintillator. Here, both the single and multi-layer perceptron neural network models have been implemented to unfold the neutron spectrum from an Am-Be neutron source. The activation function and the connectivity of the neurons have been investigated and the results have been analyzed in terms of the network's performance. The simulation results show that the neural network that utilizes the Satlins transfer function has the best performance. In addition, omitting the bias connection of the neurons improve the performance of the network. Also, the SCINFUL code is used for generating the response functions in the training phase of the process. Finally, the results of the neural network simulation have been compared with those of the FORIST unfolding code for both (241)Am-Be and (252)Cf neutron sources. The results of neural network are in good agreement with FORIST code.

A neural-network-based model for the dynamic simulation of the tire/suspension system while traversing road irregularities.

PubMed

Guarneri, Paolo; Rocca, Gianpiero; Gobbi, Massimiliano

2008-09-01

This paper deals with the simulation of the tire/suspension dynamics by using recurrent neural networks (RNNs). RNNs are derived from the multilayer feedforward neural networks, by adding feedback connections between output and input layers. The optimal network architecture derives from a parametric analysis based on the optimal tradeoff between network accuracy and size. The neural network can be trained with experimental data obtained in the laboratory from simulated road profiles (cleats). The results obtained from the neural network demonstrate good agreement with the experimental results over a wide range of operation conditions. The NN model can be effectively applied as a part of vehicle system model to accurately predict elastic bushings and tire dynamics behavior. Although the neural network model, as a black-box model, does not provide a good insight of the physical behavior of the tire/suspension system, it is a useful tool for assessing vehicle ride and noise, vibration, harshness (NVH) performance due to its good computational efficiency and accuracy.

Computational modeling of spiking neural network with learning rules from STDP and intrinsic plasticity

NASA Astrophysics Data System (ADS)

Li, Xiumin; Wang, Wei; Xue, Fangzheng; Song, Yongduan

2018-02-01

Recently there has been continuously increasing interest in building up computational models of spiking neural networks (SNN), such as the Liquid State Machine (LSM). The biologically inspired self-organized neural networks with neural plasticity can enhance the capability of computational performance, with the characteristic features of dynamical memory and recurrent connection cycles which distinguish them from the more widely used feedforward neural networks. Despite a variety of computational models for brain-like learning and information processing have been proposed, the modeling of self-organized neural networks with multi-neural plasticity is still an important open challenge. The main difficulties lie in the interplay among different forms of neural plasticity rules and understanding how structures and dynamics of neural networks shape the computational performance. In this paper, we propose a novel approach to develop the models of LSM with a biologically inspired self-organizing network based on two neural plasticity learning rules. The connectivity among excitatory neurons is adapted by spike-timing-dependent plasticity (STDP) learning; meanwhile, the degrees of neuronal excitability are regulated to maintain a moderate average activity level by another learning rule: intrinsic plasticity (IP). Our study shows that LSM with STDP+IP performs better than LSM with a random SNN or SNN obtained by STDP alone. The noticeable improvement with the proposed method is due to the better reflected competition among different neurons in the developed SNN model, as well as the more effectively encoded and processed relevant dynamic information with its learning and self-organizing mechanism. This result gives insights to the optimization of computational models of spiking neural networks with neural plasticity.

Supervision of care networks for frail community dwelling adults aged 75 years and older: protocol of a mixed methods study

PubMed Central

Verver, Didi; Merten, Hanneke; Robben, Paul; Wagner, Cordula

2015-01-01

Introduction The Dutch healthcare inspectorate (IGZ) supervises the quality and safety of healthcare in the Netherlands. Owing to the growing population of (community dwelling) older adults and changes in the Dutch healthcare system, the IGZ is exploring new methods to effectively supervise care networks that exist around frail older adults. The composition of these networks, where formal and informal care takes place, and the lack of guidelines and quality and risk indicators make supervision complicated in the current situation. Methods and analysis This study consists of four phases. The first phase identifies risks for community dwelling frail older adults in the existing literature. In the second phase, a qualitative pilot study will be conducted to assess the needs and wishes of the frail older adults concerning care and well-being, perception of risks, and the composition of their networks, collaboration and coordination between care providers involved in the network. In the third phase, questionnaires based on the results of phase II will be sent to a larger group of frail older adults (n=200) and their care providers. The results will describe the composition of their care networks and prioritise risks concerning community dwelling older adults. Also, it will provide input for the development of a new supervision framework by the IGZ. During phase IV, a second questionnaire will be sent to the participants of phase III to establish changes of perception in risks and possible changes in the care networks. The framework will be tested by the IGZ in pilots, and the researchers will evaluate these pilots and provide feedback to the IGZ. Ethics and dissemination The study protocol was approved by the Scientific Committee of the EMGO+institute and the Medical Ethical review committee of the VU University Medical Centre. Results will be presented in scientific articles and reports and at meetings. PMID:26307619

New supervised learning theory applied to cerebellar modeling for suppression of variability of saccade end points.

PubMed

Fujita, Masahiko

2013-06-01

A new supervised learning theory is proposed for a hierarchical neural network with a single hidden layer of threshold units, which can approximate any continuous transformation, and applied to a cerebellar function to suppress the end-point variability of saccades. In motor systems, feedback control can reduce noise effects if the noise is added in a pathway from a motor center to a peripheral effector; however, it cannot reduce noise effects if the noise is generated in the motor center itself: a new control scheme is necessary for such noise. The cerebellar cortex is well known as a supervised learning system, and a novel theory of cerebellar cortical function developed in this study can explain the capability of the cerebellum to feedforwardly reduce noise effects, such as end-point variability of saccades. This theory assumes that a Golgi-granule cell system can encode the strength of a mossy fiber input as the state of neuronal activity of parallel fibers. By combining these parallel fiber signals with appropriate connection weights to produce a Purkinje cell output, an arbitrary continuous input-output relationship can be obtained. By incorporating such flexible computation and learning ability in a process of saccadic gain adaptation, a new control scheme in which the cerebellar cortex feedforwardly suppresses the end-point variability when it detects a variation in saccadic commands can be devised. Computer simulation confirmed the efficiency of such learning and showed a reduction in the variability of saccadic end points, similar to results obtained from experimental data.

Plant Growth Models Using Artificial Neural Networks

NASA Technical Reports Server (NTRS)

Bubenheim, David

1997-01-01

In this paper, we descrive our motivation and approach to devloping models and the neural network architecture. Initial use of the artificial neural network for modeling the single plant process of transpiration is presented.

Artificial Neural Network for the Prediction of Chromosomal Abnormalities in Azoospermic Males.

PubMed

Akinsal, Emre Can; Haznedar, Bulent; Baydilli, Numan; Kalinli, Adem; Ozturk, Ahmet; Ekmekçioğlu, Oğuz

2018-02-04

To evaluate whether an artifical neural network helps to diagnose any chromosomal abnormalities in azoospermic males. The data of azoospermic males attending to a tertiary academic referral center were evaluated retrospectively. Height, total testicular volume, follicle stimulating hormone, luteinising hormone, total testosterone and ejaculate volume of the patients were used for the analyses. In artificial neural network, the data of 310 azoospermics were used as the education and 115 as the test set. Logistic regression analyses and discriminant analyses were performed for statistical analyses. The tests were re-analysed with a neural network. Both logistic regression analyses and artificial neural network predicted the presence or absence of chromosomal abnormalities with more than 95% accuracy. The use of artificial neural network model has yielded satisfactory results in terms of distinguishing patients whether they have any chromosomal abnormality or not.

Synchronization criteria for generalized reaction-diffusion neural networks via periodically intermittent control.

PubMed

Gan, Qintao; Lv, Tianshi; Fu, Zhenhua

2016-04-01

In this paper, the synchronization problem for a class of generalized neural networks with time-varying delays and reaction-diffusion terms is investigated concerning Neumann boundary conditions in terms of p-norm. The proposed generalized neural networks model includes reaction-diffusion local field neural networks and reaction-diffusion static neural networks as its special cases. By establishing a new inequality, some simple and useful conditions are obtained analytically to guarantee the global exponential synchronization of the addressed neural networks under the periodically intermittent control. According to the theoretical results, the influences of diffusion coefficients, diffusion space, and control rate on synchronization are analyzed. Finally, the feasibility and effectiveness of the proposed methods are shown by simulation examples, and by choosing different diffusion coefficients, diffusion spaces, and control rates, different controlled synchronization states can be obtained.

Global exponential stability of inertial memristor-based neural networks with time-varying delays and impulses.

PubMed

Zhang, Wei; Huang, Tingwen; He, Xing; Li, Chuandong

2017-11-01

In this study, we investigate the global exponential stability of inertial memristor-based neural networks with impulses and time-varying delays. We construct inertial memristor-based neural networks based on the characteristics of the inertial neural networks and memristor. Impulses with and without delays are considered when modeling the inertial neural networks simultaneously, which are of great practical significance in the current study. Some sufficient conditions are derived under the framework of the Lyapunov stability method, as well as an extended Halanay differential inequality and a new delay impulsive differential inequality, which depend on impulses with and without delays, in order to guarantee the global exponential stability of the inertial memristor-based neural networks. Finally, two numerical examples are provided to illustrate the efficiency of the proposed methods. Copyright © 2017 Elsevier Ltd. All rights reserved.

Neural joint control for Space Shuttle Remote Manipulator System

NASA Technical Reports Server (NTRS)

Atkins, Mark A.; Cox, Chadwick J.; Lothers, Michael D.; Pap, Robert M.; Thomas, Charles R.

1992-01-01

Neural networks are being used to control a robot arm in a telerobotic operation. The concept uses neural networks for both joint and inverse kinematics in a robotic control application. An upper level neural network is trained to learn inverse kinematic mappings. The output, a trajectory, is then fed to the Decentralized Adaptive Joint Controllers. This neural network implementation has shown that the controlled arm recovers from unexpected payload changes while following the reference trajectory. The neural network-based decentralized joint controller is faster, more robust and efficient than conventional approaches. Implementations of this architecture are discussed that would relax assumptions about dynamics, obstacles, and heavy loads. This system is being developed to use with the Space Shuttle Remote Manipulator System.