Optimization of Training Sets for Neural-Net Processing of Characteristic Patterns from Vibrating Solids

NASA Technical Reports Server (NTRS)

Decker, Arthur J.

2001-01-01

Artificial neural networks have been used for a number of years to process holography-generated characteristic patterns of vibrating structures. This technology depends critically on the selection and the conditioning of the training sets. A scaling operation called folding is discussed for conditioning training sets optimally for training feed-forward neural networks to process characteristic fringe patterns. Folding allows feed-forward nets to be trained easily to detect damage-induced vibration-displacement-distribution changes as small as 10 nm. A specific application to aerospace of neural-net processing of characteristic patterns is presented to motivate the conditioning and optimization effort.

Prediction of Bispectral Index during Target-controlled Infusion of Propofol and Remifentanil: A Deep Learning Approach.

PubMed

Lee, Hyung-Chul; Ryu, Ho-Geol; Chung, Eun-Jin; Jung, Chul-Woo

2018-03-01

The discrepancy between predicted effect-site concentration and measured bispectral index is problematic during intravenous anesthesia with target-controlled infusion of propofol and remifentanil. We hypothesized that bispectral index during total intravenous anesthesia would be more accurately predicted by a deep learning approach. Long short-term memory and the feed-forward neural network were sequenced to simulate the pharmacokinetic and pharmacodynamic parts of an empirical model, respectively, to predict intraoperative bispectral index during combined use of propofol and remifentanil. Inputs of long short-term memory were infusion histories of propofol and remifentanil, which were retrieved from target-controlled infusion pumps for 1,800 s at 10-s intervals. Inputs of the feed-forward network were the outputs of long short-term memory and demographic data such as age, sex, weight, and height. The final output of the feed-forward network was the bispectral index. The performance of bispectral index prediction was compared between the deep learning model and previously reported response surface model. The model hyperparameters comprised 8 memory cells in the long short-term memory layer and 16 nodes in the hidden layer of the feed-forward network. The model training and testing were performed with separate data sets of 131 and 100 cases. The concordance correlation coefficient (95% CI) were 0.561 (0.560 to 0.562) in the deep learning model, which was significantly larger than that in the response surface model (0.265 [0.263 to 0.266], P < 0.001). The deep learning model-predicted bispectral index during target-controlled infusion of propofol and remifentanil more accurately compared to the traditional model. The deep learning approach in anesthetic pharmacology seems promising because of its excellent performance and extensibility.

Negative viscosity can enhance learning of inertial dynamics.

PubMed

Huang, Felix C; Patton, James L; Mussa-Ivaldi, Ferdinando A

2009-06-01

We investigated how learning of inertial load manipulation is influenced by movement amplification with negative viscosity. Using a force-feedback device, subjects trained on anisotropic loads (5 orientations) with free movements in one of three conditions (inertia only, negative viscosity only, or combined), prior to common evaluation conditions (prescribed circular pattern with inertia only). Training with Combined-Load resulted in lower error (6.89±3.25%) compared to Inertia-Only (8.40±4.32%) and Viscosity-Only (8.17±4.13%) according to radial deviation analysis (% of trial mean radius). Combined-Load and Inertia-Only groups exhibited similar unexpected no-load trials (8.38±4.31% versus 8.91±4.70% of trial mean radius), which suggests comparable low-impedance strategies. These findings are remarkable since negative viscosity, only available during training, evidently enhanced learning when combined with inertia. Modeling analysis suggests that a feedforward after-effect of negative viscosity cannot predict such performance gains. Instead, results from Combined-Load training are consistent with greater feedforward inertia compensation along with a small increase in impedance control. The capability of the nervous system to generalize learning from negative viscosity suggests an intriguing new method for enhancing sensorimotor adaptation.

Optimization of Training Sets For Neural-Net Processing of Characteristic Patterns From Vibrating Solids

NASA Technical Reports Server (NTRS)

Decker, Arthur J. (Inventor)

2006-01-01

An artificial neural network is disclosed that processes holography generated characteristic pattern of vibrating structures along with finite-element models. The present invention provides for a folding operation for conditioning training sets for optimally training forward-neural networks to process characteristic fringe pattern. The folding pattern increases the sensitivity of the feed-forward network for detecting changes in the characteristic pattern The folding routine manipulates input pixels so as to be scaled according to the location in an intensity range rather than the position in the characteristic pattern.

A Connectionist Model of Stimulus Class Formation with a Yes/No Procedure and Compound Stimuli

ERIC Educational Resources Information Center

Tovar, Angel E.; Chavez, Alvaro Torres

2012-01-01

We analyzed stimulus class formation in a human study and in a connectionist model (CM) with a yes/no procedure, using compound stimuli. In the human study, the participants were six female undergraduate students; the CM was a feed-forward back-propagation network. Two 3-member stimulus classes were trained with a similar procedure in both the…

Feed-forward neural network model for hunger and satiety related VAS score prediction.

PubMed

Krishnan, Shaji; Hendriks, Henk F J; Hartvigsen, Merete L; de Graaf, Albert A

2016-07-07

An artificial neural network approach was chosen to model the outcome of the complex signaling pathways in the gastro-intestinal tract and other peripheral organs that eventually produce the satiety feeling in the brain upon feeding. A multilayer feed-forward neural network was trained with sets of experimental data relating concentration-time courses of plasma satiety hormones to Visual Analog Scales (VAS) scores. The network successfully predicted VAS responses from sets of satiety hormone data obtained in experiments using different food compositions. The correlation coefficients for the predicted VAS responses for test sets having i) a full set of three satiety hormones, ii) a set of only two satiety hormones, and iii) a set of only one satiety hormone were 0.96, 0.96, and 0.89, respectively. The predicted VAS responses discriminated the satiety effects of high satiating food types from less satiating food types both in orally fed and ileal infused forms. From this application of artificial neural networks, one may conclude that neural network models are very suitable to describe situations where behavior is complex and incompletely understood. However, training data sets that fit the experimental conditions need to be available.

Nonlinear adaptive inverse control via the unified model neural network

NASA Astrophysics Data System (ADS)

Jeng, Jin-Tsong; Lee, Tsu-Tian

1999-03-01

In this paper, we propose a new nonlinear adaptive inverse control via a unified model neural network. In order to overcome nonsystematic design and long training time in nonlinear adaptive inverse control, we propose the approximate transformable technique to obtain a Chebyshev Polynomials Based Unified Model (CPBUM) neural network for the feedforward/recurrent neural networks. It turns out that the proposed method can use less training time to get an inverse model. Finally, we apply this proposed method to control magnetic bearing system. The experimental results show that the proposed nonlinear adaptive inverse control architecture provides a greater flexibility and better performance in controlling magnetic bearing systems.

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.

Deterministic convergence of chaos injection-based gradient method for training feedforward neural networks.

PubMed

Zhang, Huisheng; Zhang, Ying; Xu, Dongpo; Liu, Xiaodong

2015-06-01

It has been shown that, by adding a chaotic sequence to the weight update during the training of neural networks, the chaos injection-based gradient method (CIBGM) is superior to the standard backpropagation algorithm. This paper presents the theoretical convergence analysis of CIBGM for training feedforward neural networks. We consider both the case of batch learning as well as the case of online learning. Under mild conditions, we prove the weak convergence, i.e., the training error tends to a constant and the gradient of the error function tends to zero. Moreover, the strong convergence of CIBGM is also obtained with the help of an extra condition. The theoretical results are substantiated by a simulation example.

Objective Model Selection for Identifying the Human Feedforward Response in Manual Control.

PubMed

Drop, Frank M; Pool, Daan M; van Paassen, Marinus Rene M; Mulder, Max; Bulthoff, Heinrich H

2018-01-01

Realistic manual control tasks typically involve predictable target signals and random disturbances. The human controller (HC) is hypothesized to use a feedforward control strategy for target-following, in addition to feedback control for disturbance-rejection. Little is known about human feedforward control, partly because common system identification methods have difficulty in identifying whether, and (if so) how, the HC applies a feedforward strategy. In this paper, an identification procedure is presented that aims at an objective model selection for identifying the human feedforward response, using linear time-invariant autoregressive with exogenous input models. A new model selection criterion is proposed to decide on the model order (number of parameters) and the presence of feedforward in addition to feedback. For a range of typical control tasks, it is shown by means of Monte Carlo computer simulations that the classical Bayesian information criterion (BIC) leads to selecting models that contain a feedforward path from data generated by a pure feedback model: "false-positive" feedforward detection. To eliminate these false-positives, the modified BIC includes an additional penalty on model complexity. The appropriate weighting is found through computer simulations with a hypothesized HC model prior to performing a tracking experiment. Experimental human-in-the-loop data will be considered in future work. With appropriate weighting, the method correctly identifies the HC dynamics in a wide range of control tasks, without false-positive results.

Boundedness and convergence of online gradient method with penalty for feedforward neural networks.

PubMed

Zhang, Huisheng; Wu, Wei; Liu, Fei; Yao, Mingchen

2009-06-01

In this brief, we consider an online gradient method with penalty for training feedforward neural networks. Specifically, the penalty is a term proportional to the norm of the weights. Its roles in the method are to control the magnitude of the weights and to improve the generalization performance of the network. By proving that the weights are automatically bounded in the network training with penalty, we simplify the conditions that are required for convergence of online gradient method in literature. A numerical example is given to support the theoretical analysis.

Neural Network Classifier Architectures for Phoneme Recognition. CRC Technical Note No. CRC-TN-92-001.

ERIC Educational Resources Information Center

Treurniet, William

A study applied artificial neural networks, trained with the back-propagation learning algorithm, to modelling phonemes extracted from the DARPA TIMIT multi-speaker, continuous speech data base. A number of proposed network architectures were applied to the phoneme classification task, ranging from the simple feedforward multilayer network to more…

Functional approximation using artificial neural networks in structural mechanics

NASA Technical Reports Server (NTRS)

Alam, Javed; Berke, Laszlo

1993-01-01

The artificial neural networks (ANN) methodology is an outgrowth of research in artificial intelligence. In this study, the feed-forward network model that was proposed by Rumelhart, Hinton, and Williams was applied to the mapping of functions that are encountered in structural mechanics problems. Several different network configurations were chosen to train the available data for problems in materials characterization and structural analysis of plates and shells. By using the recall process, the accuracy of these trained networks was assessed.

Inverse kinematics problem in robotics using neural networks

NASA Technical Reports Server (NTRS)

Choi, Benjamin B.; Lawrence, Charles

1992-01-01

In this paper, Multilayer Feedforward Networks are applied to the robot inverse kinematic problem. The networks are trained with endeffector position and joint angles. After training, performance is measured by having the network generate joint angles for arbitrary endeffector trajectories. A 3-degree-of-freedom (DOF) spatial manipulator is used for the study. It is found that neural networks provide a simple and effective way to both model the manipulator inverse kinematics and circumvent the problems associated with algorithmic solution methods.

The Müller-Lyer Illusion in a Computational Model of Biological Object Recognition

PubMed Central

Zeman, Astrid; Obst, Oliver; Brooks, Kevin R.; Rich, Anina N.

2013-01-01

Studying illusions provides insight into the way the brain processes information. The Müller-Lyer Illusion (MLI) is a classical geometrical illusion of size, in which perceived line length is decreased by arrowheads and increased by arrowtails. Many theories have been put forward to explain the MLI, such as misapplied size constancy scaling, the statistics of image-source relationships and the filtering properties of signal processing in primary visual areas. Artificial models of the ventral visual processing stream allow us to isolate factors hypothesised to cause the illusion and test how these affect classification performance. We trained a feed-forward feature hierarchical model, HMAX, to perform a dual category line length judgment task (short versus long) with over 90% accuracy. We then tested the system in its ability to judge relative line lengths for images in a control set versus images that induce the MLI in humans. Results from the computational model show an overall illusory effect similar to that experienced by human subjects. No natural images were used for training, implying that misapplied size constancy and image-source statistics are not necessary factors for generating the illusion. A post-hoc analysis of response weights within a representative trained network ruled out the possibility that the illusion is caused by a reliance on information at low spatial frequencies. Our results suggest that the MLI can be produced using only feed-forward, neurophysiological connections. PMID:23457510

An Indoor Positioning Technique Based on a Feed-Forward Artificial Neural Network Using Levenberg-Marquardt Learning Method

NASA Astrophysics Data System (ADS)

Pahlavani, P.; Gholami, A.; Azimi, S.

2017-09-01

This paper presents an indoor positioning technique based on a multi-layer feed-forward (MLFF) artificial neural networks (ANN). Most of the indoor received signal strength (RSS)-based WLAN positioning systems use the fingerprinting technique that can be divided into two phases: the offline (calibration) phase and the online (estimation) phase. In this paper, RSSs were collected for all references points in four directions and two periods of time (Morning and Evening). Hence, RSS readings were sampled at a regular time interval and specific orientation at each reference point. The proposed ANN based model used Levenberg-Marquardt algorithm for learning and fitting the network to the training data. This RSS readings in all references points and the known position of these references points was prepared for training phase of the proposed MLFF neural network. Eventually, the average positioning error for this network using 30% check and validation data was computed approximately 2.20 meter.

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

Design and evaluation of a Stochastic Optimal Feed-forward and Feedback Technology (SOFFT) flight control architecture

NASA Technical Reports Server (NTRS)

Ostroff, Aaron J.; Proffitt, Melissa S.

1994-01-01

This paper describes the design and evaluation of a stochastic optimal feed-forward and feedback technology (SOFFT) control architecture with emphasis on the feed-forward controller design. The SOFFT approach allows the designer to independently design the feed-forward and feedback controllers to meet separate objectives and then integrate the two controllers. The feed-forward controller has been integrated with an existing high-angle-of-attack (high-alpha) feedback controller. The feed-forward controller includes a variable command model with parameters selected to satisfy level 1 flying qualities with a high-alpha adjustment to achieve desired agility guidelines, a nonlinear interpolation approach that scales entire matrices for approximation of the plant model, and equations for calculating feed-forward gains developed for perfect plant-model tracking. The SOFFT design was applied to a nonlinear batch simulation model of an F/A-18 aircraft modified for thrust vectoring. Simulation results show that agility guidelines are met and that the SOFFT controller filters undesired pilot-induced frequencies more effectively during a tracking task than a flight controller that has the same feedback control law but does not have the SOFFT feed-forward control.

Passivity/Lyapunov based controller design for trajectory tracking of flexible joint manipulators

NASA Technical Reports Server (NTRS)

Sicard, Pierre; Wen, John T.; Lanari, Leonardo

1992-01-01

A passivity and Lyapunov based approach for the control design for the trajectory tracking problem of flexible joint robots is presented. The basic structure of the proposed controller is the sum of a model-based feedforward and a model-independent feedback. Feedforward selection and solution is analyzed for a general model for flexible joints, and for more specific and practical model structures. Passivity theory is used to design a motor state-based controller in order to input-output stabilize the error system formed by the feedforward. Observability conditions for asymptotic stability are stated and verified. In order to accommodate for modeling uncertainties and to allow for the implementation of a simplified feedforward compensation, the stability of the system is analyzed in presence of approximations in the feedforward by using a Lyapunov based robustness analysis. It is shown that under certain conditions, e.g., the desired trajectory is varying slowly enough, stability is maintained for various approximations of a canonical feedforward.

Prediction of Flow Stress in Cadmium Using Constitutive Equation and Artificial Neural Network Approach

NASA Astrophysics Data System (ADS)

Sarkar, A.; Chakravartty, J. K.

2013-10-01

A model is developed to predict the constitutive flow behavior of cadmium during compression test using artificial neural network (ANN). The inputs of the neural network are strain, strain rate, and temperature, whereas flow stress is the output. Experimental data obtained from compression tests in the temperature range -30 to 70 °C, strain range 0.1 to 0.6, and strain rate range 10-3 to 1 s-1 are employed to develop the model. A three-layer feed-forward ANN is trained with Levenberg-Marquardt training algorithm. It has been shown that the developed ANN model can efficiently and accurately predict the deformation behavior of cadmium. This trained network could predict the flow stress better than a constitutive equation of the type.

Decoding small surface codes with feedforward neural networks

NASA Astrophysics Data System (ADS)

Varsamopoulos, Savvas; Criger, Ben; Bertels, Koen

2018-01-01

Surface codes reach high error thresholds when decoded with known algorithms, but the decoding time will likely exceed the available time budget, especially for near-term implementations. To decrease the decoding time, we reduce the decoding problem to a classification problem that a feedforward neural network can solve. We investigate quantum error correction and fault tolerance at small code distances using neural network-based decoders, demonstrating that the neural network can generalize to inputs that were not provided during training and that they can reach similar or better decoding performance compared to previous algorithms. We conclude by discussing the time required by a feedforward neural network decoder in hardware.

Modelling the Flow Stress of Alloy 316L using a Multi-Layered Feed Forward Neural Network with Bayesian Regularization

NASA Astrophysics Data System (ADS)

Abiriand Bhekisipho Twala, Olufunminiyi

2017-08-01

In this paper, a multilayer feedforward neural network with Bayesian regularization constitutive model is developed for alloy 316L during high strain rate and high temperature plastic deformation. The input variables are strain rate, temperature and strain while the output value is the flow stress of the material. The results show that the use of Bayesian regularized technique reduces the potential of overfitting and overtraining. The prediction quality of the model is thereby improved. The model predictions are in good agreement with experimental measurements. The measurement data used for the network training and model comparison were taken from relevant literature. The developed model is robust as it can be generalized to deformation conditions slightly below or above the training dataset.

Learning feedback and feedforward control in a mirror-reversed visual environment.

PubMed

Kasuga, Shoko; Telgen, Sebastian; Ushiba, Junichi; Nozaki, Daichi; Diedrichsen, Jörn

2015-10-01

When we learn a novel task, the motor system needs to acquire both feedforward and feedback control. Currently, little is known about how the learning of these two mechanisms relate to each other. In the present study, we tested whether feedforward and feedback control need to be learned separately, or whether they are learned as common mechanism when a new control policy is acquired. Participants were trained to reach to two lateral and one central target in an environment with mirror (left-right)-reversed visual feedback. One group was allowed to make online movement corrections, whereas the other group only received visual information after the end of the movement. Learning of feedforward control was assessed by measuring the accuracy of the initial movement direction to lateral targets. Feedback control was measured in the responses to sudden visual perturbations of the cursor when reaching to the central target. Although feedforward control improved in both groups, it was significantly better when online corrections were not allowed. In contrast, feedback control only adaptively changed in participants who received online feedback and remained unchanged in the group without online corrections. Our findings suggest that when a new control policy is acquired, feedforward and feedback control are learned separately, and that there may be a trade-off in learning between feedback and feedforward controllers. Copyright © 2015 the American Physiological Society.

Learning feedback and feedforward control in a mirror-reversed visual environment

PubMed Central

Kasuga, Shoko; Telgen, Sebastian; Ushiba, Junichi; Nozaki, Daichi

2015-01-01

When we learn a novel task, the motor system needs to acquire both feedforward and feedback control. Currently, little is known about how the learning of these two mechanisms relate to each other. In the present study, we tested whether feedforward and feedback control need to be learned separately, or whether they are learned as common mechanism when a new control policy is acquired. Participants were trained to reach to two lateral and one central target in an environment with mirror (left-right)-reversed visual feedback. One group was allowed to make online movement corrections, whereas the other group only received visual information after the end of the movement. Learning of feedforward control was assessed by measuring the accuracy of the initial movement direction to lateral targets. Feedback control was measured in the responses to sudden visual perturbations of the cursor when reaching to the central target. Although feedforward control improved in both groups, it was significantly better when online corrections were not allowed. In contrast, feedback control only adaptively changed in participants who received online feedback and remained unchanged in the group without online corrections. Our findings suggest that when a new control policy is acquired, feedforward and feedback control are learned separately, and that there may be a trade-off in learning between feedback and feedforward controllers. PMID:26245313

Feedforward neural network model estimating pollutant removal process within mesophilic upflow anaerobic sludge blanket bioreactor treating industrial starch processing wastewater.

PubMed

Antwi, Philip; Li, Jianzheng; Meng, Jia; Deng, Kaiwen; Koblah Quashie, Frank; Li, Jiuling; Opoku Boadi, Portia

2018-06-01

In this a, three-layered feedforward-backpropagation artificial neural network (BPANN) model was developed and employed to evaluate COD removal an upflow anaerobic sludge blanket (UASB) reactor treating industrial starch processing wastewater. At the end of UASB operation, microbial community characterization revealed satisfactory composition of microbes whereas morphology depicted rod-shaped archaea. pH, COD, NH 4 + , VFA, OLR and biogas yield were selected by principal component analysis and used as input variables. Whilst tangent sigmoid function (tansig) and linear function (purelin) were assigned as activation functions at the hidden-layer and output-layer, respectively, optimum BPANN architecture was achieved with Levenberg-Marquardt algorithm (trainlm) after eleven training algorithms had been tested. Based on performance indicators such the mean squared errors, fractional variance, index of agreement and coefficient of determination (R 2 ), the BPANN model demonstrated significant performance with R 2 reaching 87%. The study revealed that, control and optimization of an anaerobic digestion process with BPANN model was feasible. Copyright © 2018 Elsevier Ltd. All rights reserved.

Strength training for a child with suspected developmental coordination disorder.

PubMed

Menz, Stacy M; Hatten, Kristin; Grant-Beuttler, Marybeth

2013-01-01

Children with developmental coordination disorder (DCD) demonstrate difficulty with feedforward motor control and use varied compensatory strategies. To examine gross motor function changes following strength training in a child with motor control difficulties. A girl aged 6 years 11 months, with apraxia and hypotonia, and demonstrating motor delays consistent with DCD. Twenty-four strength training sessions were completed using a universal exercise unit. Postintervention scores significantly improved on the Bruininks-Oseretsky test of motor proficiency, second edition, and the Canadian occupational performance measure scores and raised the developmental coordination disorder questionnaire, revised 2007, scores above the range where DCD is suspected. Nonsignificant changes in strength were observed. Improved function and significant gains in manual coordination were observed following blocked practice of isolated, simple joint movements during strength training. Improved motor skills may be because of effective use of feedforward control and improved stabilization. Strength training does not rehearse skills using momentum, explaining nonsignificant changes in locomotor or locomotion areas.

Neural-network-based state of health diagnostics for an automated radioxenon sampler/analyzer

NASA Astrophysics Data System (ADS)

Keller, Paul E.; Kangas, Lars J.; Hayes, James C.; Schrom, Brian T.; Suarez, Reynold; Hubbard, Charles W.; Heimbigner, Tom R.; McIntyre, Justin I.

2009-05-01

Artificial neural networks (ANNs) are used to determine the state-of-health (SOH) of the Automated Radioxenon Analyzer/Sampler (ARSA). ARSA is a gas collection and analysis system used for non-proliferation monitoring in detecting radioxenon released during nuclear tests. SOH diagnostics are important for automated, unmanned sensing systems so that remote detection and identification of problems can be made without onsite staff. Both recurrent and feed-forward ANNs are presented. The recurrent ANN is trained to predict sensor values based on current valve states, which control air flow, so that with only valve states the normal SOH sensor values can be predicted. Deviation between modeled value and actual is an indication of a potential problem. The feed-forward ANN acts as a nonlinear version of principal components analysis (PCA) and is trained to replicate the normal SOH sensor values. Because of ARSA's complexity, this nonlinear PCA is better able to capture the relationships among the sensors than standard linear PCA and is applicable to both sensor validation and recognizing off-normal operating conditions. Both models provide valuable information to detect impending malfunctions before they occur to avoid unscheduled shutdown. Finally, the ability of ANN methods to predict the system state is presented.

Feedforward interview technique in obstetrics and gynaecology residents: a fact or fallacy.

PubMed

Sami, Shehla; Ahmad, Amina

2015-01-01

To determine the role of Feedforward Interview (FFI) technique in motivating residents of Obstetrics and Gynaecology for better learning and performance. An explorative study with mixed method approach being employed. Department of Obstetrics and Gynaecology, Sandeman (Provincial) Hospital, Quetta, from November 2010 till May 2013. Feedforward interview technique was complimented by survey questionnaire employing similar philosophy of FFI to triangulate data through two methods. Survey questionnaire was filled-up by 21 residents and analysed by SPSS version 17. Fourteen of these participants were identified for in-depth Feedforward Interviews (FFI), based on nonprobability purposive sampling after informed consent, and content analysis was done. Feedforward interview technique enabled majority of residents in recalling minimum of 3 positive experiences, mainly related to surgical experiences, which enhanced their motivation to aspire for further improvement in this area. Hard work was the main personal contributing factor both in FFI and survey. In addition to identifying clinical experiences enhancing desire to learn, residents also reported need for more academic support as an important factor which could also boost motivation to attain better performance. Feedforward interview technique not only helps residents in recalling positive learning experiences during their training but it also has a significant influence on developing insight about one's performance and motivating residents to achieve higher academic goals.

Comparison of different modelling approaches of drive train temperature for the purposes of wind turbine failure detection

NASA Astrophysics Data System (ADS)

Tautz-Weinert, J.; Watson, S. J.

2016-09-01

Effective condition monitoring techniques for wind turbines are needed to improve maintenance processes and reduce operational costs. Normal behaviour modelling of temperatures with information from other sensors can help to detect wear processes in drive trains. In a case study, modelling of bearing and generator temperatures is investigated with operational data from the SCADA systems of more than 100 turbines. The focus is here on automated training and testing on a farm level to enable an on-line system, which will detect failures without human interpretation. Modelling based on linear combinations, artificial neural networks, adaptive neuro-fuzzy inference systems, support vector machines and Gaussian process regression is compared. The selection of suitable modelling inputs is discussed with cross-correlation analyses and a sensitivity study, which reveals that the investigated modelling techniques react in different ways to an increased number of inputs. The case study highlights advantages of modelling with linear combinations and artificial neural networks in a feedforward configuration.

Neural network feedforward control of a closed-circuit wind tunnel

NASA Astrophysics Data System (ADS)

Sutcliffe, Peter

Accurate control of wind-tunnel test conditions can be dramatically enhanced using feedforward control architectures which allow operating conditions to be maintained at a desired setpoint through the use of mathematical models as the primary source of prediction. However, as the desired accuracy of the feedforward prediction increases, the model complexity also increases, so that an ever increasing computational load is incurred. This drawback can be avoided by employing a neural network that is trained offline using the output of a high fidelity wind-tunnel mathematical model, so that the neural network can rapidly reproduce the predictions of the model with a greatly reduced computational overhead. A novel neural network database generation method, developed through the use of fractional factorial arrays, was employed such that a neural network can accurately predict wind-tunnel parameters across a wide range of operating conditions whilst trained upon a highly efficient database. The subsequent network was incorporated into a Neural Network Model Predictive Control (NNMPC) framework to allow an optimised output schedule capable of providing accurate control of the wind-tunnel operating parameters. Facilitation of an optimised path through the solution space is achieved through the use of a chaos optimisation algorithm such that a more globally optimum solution is likely to be found with less computational expense than the gradient descent method. The parameters associated with the NNMPC such as the control horizon are determined through the use of a Taguchi methodology enabling the minimum number of experiments to be carried out to determine the optimal combination. The resultant NNMPC scheme was employed upon the Hessert Low Speed Wind Tunnel at the University of Notre Dame to control the test-section temperature such that it follows a pre-determined reference trajectory during changes in the test-section velocity. Experimental testing revealed that the derived NNMPC controller provided an excellent level of control over the test-section temperature in adherence to a reference trajectory even when faced with unforeseen disturbances such as rapid changes in the operating environment.

Feedforward hysteresis compensation in trajectory control of piezoelectrically-driven nanostagers

NASA Astrophysics Data System (ADS)

Bashash, Saeid; Jalili, Nader

2006-03-01

Complex structural nonlinearities of piezoelectric materials drastically degrade their performance in variety of micro- and nano-positioning applications. From the precision positioning and control perspective, the multi-path time-history dependent hysteresis phenomenon is the most concerned nonlinearity in piezoelectric actuators to be analyzed. To realize the underlying physics of this phenomenon and to develop an efficient compensation strategy, the intelligent properties of hysteresis with the effects of non-local memories are discussed. Through performing a set of experiments on a piezoelectrically-driven nanostager with high resolution capacitive position sensor, it is shown that for the precise prediction of hysteresis path, certain memory units are required to store the previous hysteresis trajectory data. Based on the experimental observations, a constitutive memory-based mathematical modeling framework is developed and trained for the precise prediction of hysteresis path for arbitrarily assigned input profiles. Using the inverse hysteresis model, a feedforward control strategy is then developed and implemented on the nanostager to compensate for the system everpresent nonlinearity. Experimental results demonstrate that the controller remarkably eliminates the nonlinear effect if memory units are sufficiently chosen for the inverse model.

Design of Neural Networks for Fast Convergence and Accuracy

NASA Technical Reports Server (NTRS)

Maghami, Peiman G.; Sparks, Dean W., Jr.

1998-01-01

A novel procedure for the design and training of artificial neural networks, used for rapid and efficient controls and dynamics design and analysis for flexible space systems, has been developed. Artificial neural networks are employed to provide a means of evaluating the impact of design changes rapidly. Specifically, two-layer feedforward neural networks are designed to approximate the functional relationship between the component spacecraft design changes and measures of its performance. A training algorithm, based on statistical sampling theory, is presented, which guarantees that the trained networks provide a designer-specified degree of accuracy in mapping the functional relationship. Within each iteration of this statistical-based algorithm, a sequential design algorithm is used for the design and training of the feedforward network to provide rapid convergence to the network goals. Here, at each sequence a new network is trained to minimize the error of previous network. The design algorithm attempts to avoid the local minima phenomenon that hampers the traditional network training. A numerical example is performed on a spacecraft application in order to demonstrate the feasibility of the proposed approach.

Scalable learning method for feedforward neural networks using minimal-enclosing-ball approximation.

PubMed

Wang, Jun; Deng, Zhaohong; Luo, Xiaoqing; Jiang, Yizhang; Wang, Shitong

2016-06-01

Training feedforward neural networks (FNNs) is one of the most critical issues in FNNs studies. However, most FNNs training methods cannot be directly applied for very large datasets because they have high computational and space complexity. In order to tackle this problem, the CCMEB (Center-Constrained Minimum Enclosing Ball) problem in hidden feature space of FNN is discussed and a novel learning algorithm called HFSR-GCVM (hidden-feature-space regression using generalized core vector machine) is developed accordingly. In HFSR-GCVM, a novel learning criterion using L2-norm penalty-based ε-insensitive function is formulated and the parameters in the hidden nodes are generated randomly independent of the training sets. Moreover, the learning of parameters in its output layer is proved equivalent to a special CCMEB problem in FNN hidden feature space. As most CCMEB approximation based machine learning algorithms, the proposed HFSR-GCVM training algorithm has the following merits: The maximal training time of the HFSR-GCVM training is linear with the size of training datasets and the maximal space consumption is independent of the size of training datasets. The experiments on regression tasks confirm the above conclusions. Copyright © 2016 Elsevier Ltd. All rights reserved.

Comparing feed-forward versus neural gas as estimators: application to coke wastewater treatment.

PubMed

Machón-González, Iván; López-García, Hilario; Rodríguez-Iglesias, Jesús; Marañón-Maison, Elena; Castrillón-Peláez, Leonor; Fernández-Nava, Yolanda

2013-01-01

Numerous papers related to the estimation of wastewater parameters have used artificial neural networks. Although successful results have been reported, different problems have arisen such as overtraining, local minima and model instability. In this paper, two types of neural networks, feed-forward and neural gas, are trained to obtain a model that estimates the values of nitrates in the effluent stream of a three-step activated sludge system (two oxic and one anoxic). Placing the denitrification (anoxic) step at the head of the process can force denitrifying bacteria to use internal organic carbon. However, methanol has to be added to achieve high denitrification efficiencies in some industrial wastewaters. The aim of this paper is to compare the two networks in addition to suggesting a methodology to validate the models. The influence of the neighbourhood radius is important in the neural gas approach and must be selected correctly. Neural gas performs well due to its cooperation--competition procedure, with no problems of stability or overfitting arising in the experimental results. The neural gas model is also interesting for use as a direct plant model because of its robustness and deterministic behaviour.

Differences in feedforward trunk muscle activity in subgroups of patients with mechanical low back pain.

PubMed

Silfies, Sheri P; Mehta, Rupal; Smith, Sue S; Karduna, Andrew R

2009-07-01

To investigate alterations in trunk muscle timing patterns in subgroups of patients with mechanical low back pain (MLBP). Our hypothesis was that subjects with MLBP would demonstrate delayed muscle onset and have fewer muscles functioning in a feedforward manner than the control group. We further hypothesized that we would find differences between subgroups of our patients with MLBP, grouped according to diagnosis (segmental instability and noninstability). Case-control. Laboratory. Forty-three patients with chronic MLBP (25 instability, 18 noninstability) and 39 asymptomatic controls. Not applicable. Surface electromyography was used to measure onset time of 10 trunk muscles during a self-perturbation task. Trunk muscle onset latency relative to the anterior deltoid was calculated and the number of muscles functioning in feedforward determined. Activation timing patterns (P<.01; eta=.50; 1-beta=.99) and number of muscles functioning in feedforward (P=.02; eta=.30; 1-beta=.83) were statistically different between patients with MLBP and controls. The control group activated the external oblique, lumbar multifidus, and erector spinae muscles in a feedforward manner. The heterogeneous MLBP group did not activate the trunk musculature in feedforward, but responded with significantly delayed activations. MLBP subgroups demonstrated significantly different timing patterns. The noninstability MLBP subgroup activated trunk extensors in a feedforward manner, similar to the control group, but significantly earlier than the instability subgroup. Lack of feedforward activation of selected trunk musculature in patients with MLBP may result in a period of inefficient muscular stabilization. Activation timing was more impaired in the instability than the noninstability MLBP subgroup. Training specifically for recruitment timing may be an important component of the rehabilitation program.

Pulvinar thalamic nucleus allows for asynchronous spike propagation through the cortex

PubMed Central

Cortes, Nelson; van Vreeswijk, Carl

2015-01-01

We create two multilayered feedforward networks composed of excitatory and inhibitory integrate-and-fire neurons in the balanced state to investigate the role of cortico-pulvino-cortical connections. The first network consists of ten feedforward levels where a Poisson spike train with varying firing rate is applied as an input in layer one. Although the balanced state partially avoids spike synchronization during the transmission, the average firing-rate in the last layer either decays or saturates depending on the feedforward pathway gain. The last layer activity is almost independent of the input even for a carefully chosen intermediate gain. Adding connections to the feedforward pathway by a nine areas Pulvinar structure improves the firing-rate propagation to become almost linear among layers. Incoming strong pulvinar spikes balance the low feedforward gain to have a unit input-output relation in the last layer. Pulvinar neurons evoke a bimodal activity depending on the magnitude input: synchronized spike bursts between 20 and 80 Hz and an asynchronous activity for very both low and high frequency inputs. In the first regime, spikes of last feedforward layer neurons are asynchronous with weak, low frequency, oscillations in the rate. Here, the uncorrelated incoming feedforward pathway washes out the synchronized thalamic bursts. In the second regime, spikes in the whole network are asynchronous. As the number of cortical layers increases, long-range pulvinar connections can link directly two or more cortical stages avoiding their either saturation or gradual activity falling. The Pulvinar acts as a shortcut that supplies the input-output firing-rate relationship of two separated cortical areas without changing the strength of connections in the feedforward pathway. PMID:26042026

A novel and generalized approach in the inversion of geoelectrical resistivity data using Artificial Neural Networks (ANN)

NASA Astrophysics Data System (ADS)

Raj, A. Stanley; Srinivas, Y.; Oliver, D. Hudson; Muthuraj, D.

2014-03-01

The non-linear apparent resistivity problem in the subsurface study of the earth takes into account the model parameters in terms of resistivity and thickness of individual subsurface layers using the trained synthetic data by means of Artificial Neural Networks (ANN). Here we used a single layer feed-forward neural network with fast back propagation learning algorithm. So on proper training of back propagation networks it tends to give the resistivity and thickness of the subsurface layer model of the field resistivity data with reference to the synthetic data trained in the appropriate network. During training, the weights and biases of the network are iteratively adjusted to make network performance function level more efficient. On adequate training, errors are minimized and the best result is obtained using the artificial neural networks. The network is trained with more number of VES data and this trained network is demonstrated by the field data. The accuracy of inversion depends upon the number of data trained. In this novel and specially designed algorithm, the interpretation of the vertical electrical sounding has been done successfully with the more accurate layer model.

Design of Neural Networks for Fast Convergence and Accuracy: Dynamics and Control

NASA Technical Reports Server (NTRS)

Maghami, Peiman G.; Sparks, Dean W., Jr.

1997-01-01

A procedure for the design and training of artificial neural networks, used for rapid and efficient controls and dynamics design and analysis for flexible space systems, has been developed. Artificial neural networks are employed, such that once properly trained, they provide a means of evaluating the impact of design changes rapidly. Specifically, two-layer feedforward neural networks are designed to approximate the functional relationship between the component/spacecraft design changes and measures of its performance or nonlinear dynamics of the system/components. A training algorithm, based on statistical sampling theory, is presented, which guarantees that the trained networks provide a designer-specified degree of accuracy in mapping the functional relationship. Within each iteration of this statistical-based algorithm, a sequential design algorithm is used for the design and training of the feedforward network to provide rapid convergence to the network goals. Here, at each sequence a new network is trained to minimize the error of previous network. The proposed method should work for applications wherein an arbitrary large source of training data can be generated. Two numerical examples are performed on a spacecraft application in order to demonstrate the feasibility of the proposed approach.

Design of neural networks for fast convergence and accuracy: dynamics and control.

PubMed

Maghami, P G; Sparks, D R

2000-01-01

A procedure for the design and training of artificial neural networks, used for rapid and efficient controls and dynamics design and analysis for flexible space systems, has been developed. Artificial neural networks are employed, such that once properly trained, they provide a means of evaluating the impact of design changes rapidly. Specifically, two-layer feedforward neural networks are designed to approximate the functional relationship between the component/spacecraft design changes and measures of its performance or nonlinear dynamics of the system/components. A training algorithm, based on statistical sampling theory, is presented, which guarantees that the trained networks provide a designer-specified degree of accuracy in mapping the functional relationship. Within each iteration of this statistical-based algorithm, a sequential design algorithm is used for the design and training of the feedforward network to provide rapid convergence to the network goals. Here, at each sequence a new network is trained to minimize the error of previous network. The proposed method should work for applications wherein an arbitrary large source of training data can be generated. Two numerical examples are performed on a spacecraft application in order to demonstrate the feasibility of the proposed approach.

Neural control and transient analysis of the LCL-type resonant converter

NASA Astrophysics Data System (ADS)

Zouggar, S.; Nait Charif, H.; Azizi, M.

2000-07-01

This paper proposes a generalised inverse learning structure to control the LCL converter. A feedforward neural network is trained to act as an inverse model of the LCL converter then both are cascaded such that the composed system results in an identity mapping between desired response and the LCL output voltage. Using the large signal model, we analyse the transient output response of the controlled LCL converter in the case of large variation of the load. The simulation results show the efficiency of using neural networks to regulate the LCL converter.

Shared internal models for feedforward and feedback control.

PubMed

Wagner, Mark J; Smith, Maurice A

2008-10-15

A child often learns to ride a bicycle in the driveway, free of unforeseen obstacles. Yet when she first rides in the street, we hope that if a car suddenly pulls out in front of her, she will combine her innate goal of avoiding an accident with her learned knowledge of the bicycle, and steer away or brake. In general, when we train to perform a new motor task, our learning is most robust if it updates the rules of online error correction to reflect the rules and goals of the new task. Here we provide direct evidence that, after a new feedforward motor adaptation, motor feedback responses to unanticipated errors become precisely task appropriate, even when such errors were never experienced during training. To study this ability, we asked how, if at all, do online responses to occasional, unanticipated force pulses during reaching arm movements change after adapting to altered arm dynamics? Specifically, do they change in a task-appropriate manner? In our task, subjects learned novel velocity-dependent dynamics. However, occasional force-pulse perturbations produced unanticipated changes in velocity. Therefore, after adaptation, task-appropriate responses to unanticipated pulses should compensate corresponding changes in velocity-dependent dynamics. We found that after adaptation, pulse responses precisely compensated these changes, although they were never trained to do so. These results provide evidence for a smart feedback controller which automatically produces responses specific to the learned dynamics of the current task. To accomplish this, the neural processes underlying feedback control must (1) be capable of accurate real-time state prediction for velocity via a forward model and (2) have access to recently learned changes in internal models of limb dynamics.

Combined Effects of Feedforward Inhibition and Excitation in Thalamocortical Circuit on the Transitions of Epileptic Seizures

PubMed Central

Fan, Denggui; Duan, Lixia; Wang, Qian; Luan, Guoming

2017-01-01

The mechanisms underlying electrophysiologically observed two-way transitions between absence and tonic-clonic epileptic seizures in cerebral cortex remain unknown. The interplay within thalamocortical network is believed to give rise to these epileptic multiple modes of activity and transitions between them. In particular, it is thought that in some areas of cortex there exists feedforward inhibition from specific relay nucleus of thalamus (TC) to inhibitory neuronal population (IN) which has even more stronger functions on cortical activities than the known feedforward excitation from TC to excitatory neuronal population (EX). Inspired by this, we proposed a modified computational model by introducing feedforward inhibitory connectivity within thalamocortical circuit, to systematically investigate the combined effects of feedforward inhibition and excitation on transitions of epileptic seizures. We first found that the feedforward excitation can induce the transition from tonic oscillation to spike and wave discharges (SWD) in cortex, i.e., the epileptic tonic-absence seizures, with the fixed weak feedforward inhibition. Thereinto, the phase of absence seizures corresponding to strong feedforward excitation can be further transformed into the clonic oscillations with the increasing of feedforward inhibition, representing the epileptic absence-clonic seizures. We also observed the other fascinating dynamical states, such as periodic 2/3/4-spike and wave discharges, reversed SWD and clonic oscillations, as well as saturated firings. More importantly, we can identify the stable parameter regions representing the tonic-clonic oscillations and SWD discharges of epileptic seizures on the 2-D plane composed of feedforward inhibition and excitation, where the physiologically plausible transition pathways between tonic-clonic and absence seizures can be figured out. These results indicate the functional role of feedforward pathways in controlling epileptic seizures and the modified thalamocortical model may provide a guide for future efforts to mechanistically link feedforward pathways in the pathogenesis of epileptic seizures. PMID:28736520

Weight-elimination neural networks applied to coronary surgery mortality prediction.

PubMed

Ennett, Colleen M; Frize, Monique

2003-06-01

The objective was to assess the effectiveness of the weight-elimination cost function in improving classification performance of artificial neural networks (ANNs) and to observe how changing the a priori distribution of the training set affects network performance. Backpropagation feedforward ANNs with and without weight-elimination estimated mortality for coronary artery surgery patients. The ANNs were trained and tested on cases with 32 input variables describing the patient's medical history; the output variable was in-hospital mortality (mortality rates: training 3.7%, test 3.8%). Artificial training sets with mortality rates of 20%, 50%, and 80% were created to observe the impact of training with a higher-than-normal prevalence. When the results were averaged, weight-elimination networks achieved higher sensitivity rates than those without weight-elimination. Networks trained on higher-than-normal prevalence achieved higher sensitivity rates at the cost of lower specificity and correct classification. The weight-elimination cost function can improve the classification performance when the network is trained with a higher-than-normal prevalence. A network trained with a moderately high artificial mortality rate (artificial mortality rate of 20%) can improve the sensitivity of the model without significantly affecting other aspects of the model's performance. The ANN mortality model achieved comparable performance as additive and statistical models for coronary surgery mortality estimation in the literature.

Theory of cortical function

PubMed Central

Heeger, David J.

2017-01-01

Most models of sensory processing in the brain have a feedforward architecture in which each stage comprises simple linear filtering operations and nonlinearities. Models of this form have been used to explain a wide range of neurophysiological and psychophysical data, and many recent successes in artificial intelligence (with deep convolutional neural nets) are based on this architecture. However, neocortex is not a feedforward architecture. This paper proposes a first step toward an alternative computational framework in which neural activity in each brain area depends on a combination of feedforward drive (bottom-up from the previous processing stage), feedback drive (top-down context from the next stage), and prior drive (expectation). The relative contributions of feedforward drive, feedback drive, and prior drive are controlled by a handful of state parameters, which I hypothesize correspond to neuromodulators and oscillatory activity. In some states, neural responses are dominated by the feedforward drive and the theory is identical to a conventional feedforward model, thereby preserving all of the desirable features of those models. In other states, the theory is a generative model that constructs a sensory representation from an abstract representation, like memory recall. In still other states, the theory combines prior expectation with sensory input, explores different possible perceptual interpretations of ambiguous sensory inputs, and predicts forward in time. The theory, therefore, offers an empirically testable framework for understanding how the cortex accomplishes inference, exploration, and prediction. PMID:28167793

Feed-Forward Propagation of Temporal and Rate Information between Cortical Populations during Coherent Activation in Engineered In Vitro Networks.

PubMed

DeMarse, Thomas B; Pan, Liangbin; Alagapan, Sankaraleengam; Brewer, Gregory J; Wheeler, Bruce C

2016-01-01

Transient propagation of information across neuronal assembles is thought to underlie many cognitive processes. However, the nature of the neural code that is embedded within these transmissions remains uncertain. Much of our understanding of how information is transmitted among these assemblies has been derived from computational models. While these models have been instrumental in understanding these processes they often make simplifying assumptions about the biophysical properties of neurons that may influence the nature and properties expressed. To address this issue we created an in vitro analog of a feed-forward network composed of two small populations (also referred to as assemblies or layers) of living dissociated rat cortical neurons. The populations were separated by, and communicated through, a microelectromechanical systems (MEMS) device containing a strip of microscale tunnels. Delayed culturing of one population in the first layer followed by the second a few days later induced the unidirectional growth of axons through the microtunnels resulting in a primarily feed-forward communication between these two small neural populations. In this study we systematically manipulated the number of tunnels that connected each layer and hence, the number of axons providing communication between those populations. We then assess the effect of reducing the number of tunnels has upon the properties of between-layer communication capacity and fidelity of neural transmission among spike trains transmitted across and within layers. We show evidence based on Victor-Purpura's and van Rossum's spike train similarity metrics supporting the presence of both rate and temporal information embedded within these transmissions whose fidelity increased during communication both between and within layers when the number of tunnels are increased. We also provide evidence reinforcing the role of synchronized activity upon transmission fidelity during the spontaneous synchronized network burst events that propagated between layers and highlight the potential applications of these MEMs devices as a tool for further investigation of structure and functional dynamics among neural populations.

Feed-Forward Propagation of Temporal and Rate Information between Cortical Populations during Coherent Activation in Engineered In Vitro Networks

PubMed Central

DeMarse, Thomas B.; Pan, Liangbin; Alagapan, Sankaraleengam; Brewer, Gregory J.; Wheeler, Bruce C.

2016-01-01

Transient propagation of information across neuronal assembles is thought to underlie many cognitive processes. However, the nature of the neural code that is embedded within these transmissions remains uncertain. Much of our understanding of how information is transmitted among these assemblies has been derived from computational models. While these models have been instrumental in understanding these processes they often make simplifying assumptions about the biophysical properties of neurons that may influence the nature and properties expressed. To address this issue we created an in vitro analog of a feed-forward network composed of two small populations (also referred to as assemblies or layers) of living dissociated rat cortical neurons. The populations were separated by, and communicated through, a microelectromechanical systems (MEMS) device containing a strip of microscale tunnels. Delayed culturing of one population in the first layer followed by the second a few days later induced the unidirectional growth of axons through the microtunnels resulting in a primarily feed-forward communication between these two small neural populations. In this study we systematically manipulated the number of tunnels that connected each layer and hence, the number of axons providing communication between those populations. We then assess the effect of reducing the number of tunnels has upon the properties of between-layer communication capacity and fidelity of neural transmission among spike trains transmitted across and within layers. We show evidence based on Victor-Purpura’s and van Rossum’s spike train similarity metrics supporting the presence of both rate and temporal information embedded within these transmissions whose fidelity increased during communication both between and within layers when the number of tunnels are increased. We also provide evidence reinforcing the role of synchronized activity upon transmission fidelity during the spontaneous synchronized network burst events that propagated between layers and highlight the potential applications of these MEMs devices as a tool for further investigation of structure and functional dynamics among neural populations. PMID:27147977

Identification of the feedforward component in manual control with predictable target signals.

PubMed

Drop, Frank M; Pool, Daan M; Damveld, Herman J; van Paassen, Marinus M; Mulder, Max

2013-12-01

In the manual control of a dynamic system, the human controller (HC) often follows a visible and predictable reference path. Compared with a purely feedback control strategy, performance can be improved by making use of this knowledge of the reference. The operator could effectively introduce feedforward control in conjunction with a feedback path to compensate for errors, as hypothesized in literature. However, feedforward behavior has never been identified from experimental data, nor have the hypothesized models been validated. This paper investigates human control behavior in pursuit tracking of a predictable reference signal while being perturbed by a quasi-random multisine disturbance signal. An experiment was done in which the relative strength of the target and disturbance signals were systematically varied. The anticipated changes in control behavior were studied by means of an ARX model analysis and by fitting three parametric HC models: two different feedback models and a combined feedforward and feedback model. The ARX analysis shows that the experiment participants employed control action on both the error and the target signal. The control action on the target was similar to the inverse of the system dynamics. Model fits show that this behavior can be modeled best by the combined feedforward and feedback model.

Motor control and the management of musculoskeletal dysfunction.

PubMed

van Vliet, Paulette M; Heneghan, Nicola R

2006-08-01

This paper aims to develop understanding of three important motor control issues--feedforward mechanisms, cortical plasticity and task-specificity and assess the implications for musculoskeletal practice. A model of control for the reach-to-grasp movement illustrates how the central nervous system integrates sensorimotor processes to control complex movements. Feedforward mechanisms, an essential element of motor control, are altered in neurologically intact patients with chronic neck pain and low back pain. In healthy subjects, cortical mapping studies using transcranial magnetic stimulation have demonstrated that neural pathways adapt according to what and how much is practised. Neuroplasticity has also been demonstrated in a number of musculoskeletal conditions, where cortical maps are altered compared to normal. Behavioural and neurophysiological studies indicate that environmental and task constraints such as the goal of the task and an object's shape and size, are determinants of the motor schema for reaching and other movements. Consideration of motor control issues as well as signs and symptoms, may facilitate management of musculoskeletal conditions and improve outcome. Practice of entire everyday tasks at an early stage and systematic variation of the task is recommended. Training should be directed with the aim of re-educating feedforward mechanisms where necessary and the amount of practice should be sufficient to cause changes in cortical activity.

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.

Cellular automata simulation of topological effects on the dynamics of feed-forward motifs

PubMed Central

Apte, Advait A; Cain, John W; Bonchev, Danail G; Fong, Stephen S

2008-01-01

Background Feed-forward motifs are important functional modules in biological and other complex networks. The functionality of feed-forward motifs and other network motifs is largely dictated by the connectivity of the individual network components. While studies on the dynamics of motifs and networks are usually devoted to the temporal or spatial description of processes, this study focuses on the relationship between the specific architecture and the overall rate of the processes of the feed-forward family of motifs, including double and triple feed-forward loops. The search for the most efficient network architecture could be of particular interest for regulatory or signaling pathways in biology, as well as in computational and communication systems. Results Feed-forward motif dynamics were studied using cellular automata and compared with differential equation modeling. The number of cellular automata iterations needed for a 100% conversion of a substrate into a target product was used as an inverse measure of the transformation rate. Several basic topological patterns were identified that order the specific feed-forward constructions according to the rate of dynamics they enable. At the same number of network nodes and constant other parameters, the bi-parallel and tri-parallel motifs provide higher network efficacy than single feed-forward motifs. Additionally, a topological property of isodynamicity was identified for feed-forward motifs where different network architectures resulted in the same overall rate of the target production. Conclusion It was shown for classes of structural motifs with feed-forward architecture that network topology affects the overall rate of a process in a quantitatively predictable manner. These fundamental results can be used as a basis for simulating larger networks as combinations of smaller network modules with implications on studying synthetic gene circuits, small regulatory systems, and eventually dynamic whole-cell models. PMID:18304325

Effects of controlled element dynamics on human feedforward behavior in ramp-tracking tasks.

PubMed

Laurense, Vincent A; Pool, Daan M; Damveld, Herman J; van Paassen, Marinus René M; Mulder, Max

2015-02-01

In real-life manual control tasks, human controllers are often required to follow a visible and predictable reference signal, enabling them to use feedforward control actions in conjunction with feedback actions that compensate for errors. Little is known about human control behavior in these situations. This paper investigates how humans adapt their feedforward control dynamics to the controlled element dynamics in a combined ramp-tracking and disturbance-rejection task. A human-in-the-loop experiment is performed with a pursuit display and vehicle-like controlled elements, ranging from a single integrator through second-order systems with a break frequency at either 3, 2, or 1 rad/s, to a double integrator. Because the potential benefits of feedforward control increase with steeper ramp segments in the target signal, three steepness levels are tested to investigate their possible effect on feedforward control with the various controlled elements. Analyses with four novel models of the operator, fitted to time-domain data, reveal feedforward control for all tested controlled elements and both (nonzero) tested levels of ramp steepness. For the range of controlled element dynamics investigated, it is found that humans adapt to these dynamics in their feedforward response, with a close to perfect inversion of the controlled element dynamics. No significant effects of ramp steepness on the feedforward model parameters are found.

Learning a single-hidden layer feedforward neural network using a rank correlation-based strategy with application to high dimensional gene expression and proteomic spectra datasets in cancer detection.

PubMed

Belciug, Smaranda; Gorunescu, Florin

2018-06-08

Methods based on microarrays (MA), mass spectrometry (MS), and machine learning (ML) algorithms have evolved rapidly in recent years, allowing for early detection of several types of cancer. A pitfall of these approaches, however, is the overfitting of data due to large number of attributes and small number of instances -- a phenomenon known as the 'curse of dimensionality'. A potentially fruitful idea to avoid this drawback is to develop algorithms that combine fast computation with a filtering module for the attributes. The goal of this paper is to propose a statistical strategy to initiate the hidden nodes of a single-hidden layer feedforward neural network (SLFN) by using both the knowledge embedded in data and a filtering mechanism for attribute relevance. In order to attest its feasibility, the proposed model has been tested on five publicly available high-dimensional datasets: breast, lung, colon, and ovarian cancer regarding gene expression and proteomic spectra provided by cDNA arrays, DNA microarray, and MS. The novel algorithm, called adaptive SLFN (aSLFN), has been compared with four major classification algorithms: traditional ELM, radial basis function network (RBF), single-hidden layer feedforward neural network trained by backpropagation algorithm (BP-SLFN), and support vector-machine (SVM). Experimental results showed that the classification performance of aSLFN is competitive with the comparison models. Copyright © 2018. Published by Elsevier Inc.

Mirrored STDP Implements Autoencoder Learning in a Network of Spiking Neurons.

PubMed

Burbank, Kendra S

2015-12-01

The autoencoder algorithm is a simple but powerful unsupervised method for training neural networks. Autoencoder networks can learn sparse distributed codes similar to those seen in cortical sensory areas such as visual area V1, but they can also be stacked to learn increasingly abstract representations. Several computational neuroscience models of sensory areas, including Olshausen & Field's Sparse Coding algorithm, can be seen as autoencoder variants, and autoencoders have seen extensive use in the machine learning community. Despite their power and versatility, autoencoders have been difficult to implement in a biologically realistic fashion. The challenges include their need to calculate differences between two neuronal activities and their requirement for learning rules which lead to identical changes at feedforward and feedback connections. Here, we study a biologically realistic network of integrate-and-fire neurons with anatomical connectivity and synaptic plasticity that closely matches that observed in cortical sensory areas. Our choice of synaptic plasticity rules is inspired by recent experimental and theoretical results suggesting that learning at feedback connections may have a different form from learning at feedforward connections, and our results depend critically on this novel choice of plasticity rules. Specifically, we propose that plasticity rules at feedforward versus feedback connections are temporally opposed versions of spike-timing dependent plasticity (STDP), leading to a symmetric combined rule we call Mirrored STDP (mSTDP). We show that with mSTDP, our network follows a learning rule that approximately minimizes an autoencoder loss function. When trained with whitened natural image patches, the learned synaptic weights resemble the receptive fields seen in V1. Our results use realistic synaptic plasticity rules to show that the powerful autoencoder learning algorithm could be within the reach of real biological networks.

Mirrored STDP Implements Autoencoder Learning in a Network of Spiking Neurons

PubMed Central

Burbank, Kendra S.

2015-01-01

The autoencoder algorithm is a simple but powerful unsupervised method for training neural networks. Autoencoder networks can learn sparse distributed codes similar to those seen in cortical sensory areas such as visual area V1, but they can also be stacked to learn increasingly abstract representations. Several computational neuroscience models of sensory areas, including Olshausen & Field’s Sparse Coding algorithm, can be seen as autoencoder variants, and autoencoders have seen extensive use in the machine learning community. Despite their power and versatility, autoencoders have been difficult to implement in a biologically realistic fashion. The challenges include their need to calculate differences between two neuronal activities and their requirement for learning rules which lead to identical changes at feedforward and feedback connections. Here, we study a biologically realistic network of integrate-and-fire neurons with anatomical connectivity and synaptic plasticity that closely matches that observed in cortical sensory areas. Our choice of synaptic plasticity rules is inspired by recent experimental and theoretical results suggesting that learning at feedback connections may have a different form from learning at feedforward connections, and our results depend critically on this novel choice of plasticity rules. Specifically, we propose that plasticity rules at feedforward versus feedback connections are temporally opposed versions of spike-timing dependent plasticity (STDP), leading to a symmetric combined rule we call Mirrored STDP (mSTDP). We show that with mSTDP, our network follows a learning rule that approximately minimizes an autoencoder loss function. When trained with whitened natural image patches, the learned synaptic weights resemble the receptive fields seen in V1. Our results use realistic synaptic plasticity rules to show that the powerful autoencoder learning algorithm could be within the reach of real biological networks. PMID:26633645

Real-Time Monitoring and Fault Diagnosis of a Low Power Hub Motor Using Feedforward Neural Network.

PubMed

Şimşir, Mehmet; Bayır, Raif; Uyaroğlu, Yılmaz

2016-01-01

Low power hub motors are widely used in electromechanical systems such as electrical bicycles and solar vehicles due to their robustness and compact structure. Such systems driven by hub motors (in wheel motors) encounter previously defined and undefined faults under operation. It may inevitably lead to the interruption of the electromechanical system operation; hence, economic losses take place at certain times. Therefore, in order to maintain system operation sustainability, the motor should be precisely monitored and the faults are diagnosed considering various significant motor parameters. In this study, the artificial feedforward backpropagation neural network approach is proposed to real-time monitor and diagnose the faults of the hub motor by measuring seven main system parameters. So as to construct a necessary model, we trained the model, using a data set consisting of 4160 samples where each has 7 parameters, by the MATLAB environment until the best model is obtained. The results are encouraging and meaningful for the specific motor and the developed model may be applicable to other types of hub motors. The prosperous model of the whole system was embedded into Arduino Due microcontroller card and the mobile real-time monitoring and fault diagnosis system prototype for hub motor was designed and manufactured.

Real-Time Monitoring and Fault Diagnosis of a Low Power Hub Motor Using Feedforward Neural Network

PubMed Central

Şimşir, Mehmet; Bayır, Raif; Uyaroğlu, Yılmaz

2016-01-01

Low power hub motors are widely used in electromechanical systems such as electrical bicycles and solar vehicles due to their robustness and compact structure. Such systems driven by hub motors (in wheel motors) encounter previously defined and undefined faults under operation. It may inevitably lead to the interruption of the electromechanical system operation; hence, economic losses take place at certain times. Therefore, in order to maintain system operation sustainability, the motor should be precisely monitored and the faults are diagnosed considering various significant motor parameters. In this study, the artificial feedforward backpropagation neural network approach is proposed to real-time monitor and diagnose the faults of the hub motor by measuring seven main system parameters. So as to construct a necessary model, we trained the model, using a data set consisting of 4160 samples where each has 7 parameters, by the MATLAB environment until the best model is obtained. The results are encouraging and meaningful for the specific motor and the developed model may be applicable to other types of hub motors. The prosperous model of the whole system was embedded into Arduino Due microcontroller card and the mobile real-time monitoring and fault diagnosis system prototype for hub motor was designed and manufactured. PMID:26819590

Unorganized machines for seasonal streamflow series forecasting.

PubMed

Siqueira, Hugo; Boccato, Levy; Attux, Romis; Lyra, Christiano

2014-05-01

Modern unorganized machines--extreme learning machines and echo state networks--provide an elegant balance between processing capability and mathematical simplicity, circumventing the difficulties associated with the conventional training approaches of feedforward/recurrent neural networks (FNNs/RNNs). This work performs a detailed investigation of the applicability of unorganized architectures to the problem of seasonal streamflow series forecasting, considering scenarios associated with four Brazilian hydroelectric plants and four distinct prediction horizons. Experimental results indicate the pertinence of these models to the focused task.

Modeling the low-velocity impact characteristics of woven glass epoxy composite laminates using artificial neural networks

NASA Astrophysics Data System (ADS)

Mathivanan, N. Rajesh; Mouli, Chandra

2012-12-01

In this work, a new methodology based on artificial neural networks (ANN) has been developed to study the low-velocity impact characteristics of woven glass epoxy laminates of EP3 grade. To train and test the networks, multiple impact cases have been generated using statistical analysis of variance (ANOVA). Experimental tests were performed using an instrumented falling-weight impact-testing machine. Different impact velocities and impact energies on different thicknesses of laminates were considered as the input parameters of the ANN model. This model is a feed-forward back-propagation neural network. Using the input/output data of the experiments, the model was trained and tested. Further, the effects of the low-velocity impact response of the laminates at different energy levels were investigated by studying the cause-effect relationship among the influential factors using response surface methodology. The most significant parameter is determined from the other input variables through ANOVA.

Bringing Interpretability and Visualization with Artificial Neural Networks

ERIC Educational Resources Information Center

Gritsenko, Andrey

2017-01-01

Extreme Learning Machine (ELM) is a training algorithm for Single-Layer Feed-forward Neural Network (SLFN). The difference in theory of ELM from other training algorithms is in the existence of explicitly-given solution due to the immutability of initialed weights. In practice, ELMs achieve performance similar to that of other state-of-the-art…

Feedforward Inhibition and Synaptic Scaling – Two Sides of the Same Coin?

PubMed Central

Lücke, Jörg

2012-01-01

Feedforward inhibition and synaptic scaling are important adaptive processes that control the total input a neuron can receive from its afferents. While often studied in isolation, the two have been reported to co-occur in various brain regions. The functional implications of their interactions remain unclear, however. Based on a probabilistic modeling approach, we show here that fast feedforward inhibition and synaptic scaling interact synergistically during unsupervised learning. In technical terms, we model the input to a neural circuit using a normalized mixture model with Poisson noise. We demonstrate analytically and numerically that, in the presence of lateral inhibition introducing competition between different neurons, Hebbian plasticity and synaptic scaling approximate the optimal maximum likelihood solutions for this model. Our results suggest that, beyond its conventional use as a mechanism to remove undesired pattern variations, input normalization can make typical neural interaction and learning rules optimal on the stimulus subspace defined through feedforward inhibition. Furthermore, learning within this subspace is more efficient in practice, as it helps avoid locally optimal solutions. Our results suggest a close connection between feedforward inhibition and synaptic scaling which may have important functional implications for general cortical processing. PMID:22457610

Feedforward inhibition and synaptic scaling--two sides of the same coin?

PubMed

Keck, Christian; Savin, Cristina; Lücke, Jörg

2012-01-01

Feedforward inhibition and synaptic scaling are important adaptive processes that control the total input a neuron can receive from its afferents. While often studied in isolation, the two have been reported to co-occur in various brain regions. The functional implications of their interactions remain unclear, however. Based on a probabilistic modeling approach, we show here that fast feedforward inhibition and synaptic scaling interact synergistically during unsupervised learning. In technical terms, we model the input to a neural circuit using a normalized mixture model with Poisson noise. We demonstrate analytically and numerically that, in the presence of lateral inhibition introducing competition between different neurons, Hebbian plasticity and synaptic scaling approximate the optimal maximum likelihood solutions for this model. Our results suggest that, beyond its conventional use as a mechanism to remove undesired pattern variations, input normalization can make typical neural interaction and learning rules optimal on the stimulus subspace defined through feedforward inhibition. Furthermore, learning within this subspace is more efficient in practice, as it helps avoid locally optimal solutions. Our results suggest a close connection between feedforward inhibition and synaptic scaling which may have important functional implications for general cortical processing.

Application of DBNs for concerned internet information detecting

NASA Astrophysics Data System (ADS)

Wang, Yanfang; Gao, Song

2017-03-01

In recent years, deep learning has achieved great success in many fields, ranging from voice recognition and image classification to computer vision. In this study we apply DBNs to concerned internet information in Chinese detecting problem, since there are inherent differences between English and Chinese. Contrastive divergence (CD) is employed in the DBNs to learn a multi-layer generative model from numerous unlabeled data. The features obtained by this model are used to initialize the feed-forward neural network, which can be fine-tuned with backpropagation. Experiment results indicate that, the model and training method we proposed can be used to detect the concerned internet information effectively and accurately.

Accelerated Training for Large Feedforward Neural Networks

NASA Technical Reports Server (NTRS)

Stepniewski, Slawomir W.; Jorgensen, Charles C.

1998-01-01

In this paper we introduce a new training algorithm, the scaled variable metric (SVM) method. Our approach attempts to increase the convergence rate of the modified variable metric method. It is also combined with the RBackprop algorithm, which computes the product of the matrix of second derivatives (Hessian) with an arbitrary vector. The RBackprop method allows us to avoid computationally expensive, direct line searches. In addition, it can be utilized in the new, 'predictive' updating technique of the inverse Hessian approximation. We have used directional slope testing to adjust the step size and found that this strategy works exceptionally well in conjunction with the Rbackprop algorithm. Some supplementary, but nevertheless important enhancements to the basic training scheme such as improved setting of a scaling factor for the variable metric update and computationally more efficient procedure for updating the inverse Hessian approximation are presented as well. We summarize by comparing the SVM method with four first- and second- order optimization algorithms including a very effective implementation of the Levenberg-Marquardt method. Our tests indicate promising computational speed gains of the new training technique, particularly for large feedforward networks, i.e., for problems where the training process may be the most laborious.

Deep learning beyond Lefschetz thimbles

NASA Astrophysics Data System (ADS)

Alexandru, Andrei; Bedaque, Paulo F.; Lamm, Henry; Lawrence, Scott

2017-11-01

The generalized thimble method to treat field theories with sign problems requires repeatedly solving the computationally expensive holomorphic flow equations. We present a machine learning technique to bypass this problem. The central idea is to obtain a few field configurations via the flow equations to train a feed-forward neural network. The trained network defines a new manifold of integration which reduces the sign problem and can be rapidly sampled. We present results for the 1 +1 dimensional Thirring model with Wilson fermions on sizable lattices. In addition to the gain in speed, the parametrization of the integration manifold we use avoids the "trapping" of Monte Carlo chains which plagues large-flow calculations, a considerable shortcoming of the previous attempts.

Degraded expression of learned feedforward control in movements released by startle.

PubMed

Wright, Zachary A; Carlsen, Anthony N; MacKinnon, Colum D; Patton, James L

2015-08-01

Recent work has shown that preplanned motor programs can be rapidly released via fast conducting pathways using a startling acoustic stimulus. Our question was whether the startle-elicited response might also release a recently learned internal model, which draws on experience to predict and compensate for expected perturbations in a feedforward manner. Our initial investigation using adaptation to robotically produced forces showed some evidence of this, but the results were potentially confounded by co-contraction caused by startle. In this study, we eliminated this confound by asking subjects to make reaching movements in the presence of a visual distortion. Results show that a startle stimulus (1) decreased performance of the recently learned task and (2) reduced after-effect magnitude. Since the recall of learned control was reduced, but not eliminated during startle trials, we suggest that multiple neural centers (cortical and subcortical) are involved in such learning and adaptation. These findings have implications for motor training in areas such as piloting, teleoperation, sports, and rehabilitation.

Temporal neural networks and transient analysis of complex engineering systems

NASA Astrophysics Data System (ADS)

Uluyol, Onder

A theory is introduced for a multi-layered Local Output Gamma Feedback (LOGF) neural network within the paradigm of Locally-Recurrent Globally-Feedforward neural networks. It is developed for the identification, prediction, and control tasks of spatio-temporal systems and allows for the presentation of different time scales through incorporation of a gamma memory. It is initially applied to the tasks of sunspot and Mackey-Glass series prediction as benchmarks, then it is extended to the task of power level control of a nuclear reactor at different fuel cycle conditions. The developed LOGF neuron model can also be viewed as a Transformed Input and State (TIS) Gamma memory for neural network architectures for temporal processing. The novel LOGF neuron model extends the static neuron model by incorporating into it a short-term memory structure in the form of a digital gamma filter. A feedforward neural network made up of LOGF neurons can thus be used to model dynamic systems. A learning algorithm based upon the Backpropagation-Through-Time (BTT) approach is derived. It is applicable for training a general L-layer LOGF neural network. The spatial and temporal weights and parameters of the network are iteratively optimized for a given problem using the derived learning algorithm.

Input filter compensation for switching regulators

NASA Technical Reports Server (NTRS)

Lee, F. C.

1984-01-01

Problems caused by input filter interaction and conventional input filter design techniques are discussed. The concept of feedforward control is modeled with an input filter and a buck regulator. Experimental measurement and comparison to the analytical predictions is carried out. Transient response and the use of a feedforward loop to stabilize the regulator system is described. Other possible applications for feedforward control are included.

Feed-forward segmentation of figure-ground and assignment of border-ownership.

PubMed

Supèr, Hans; Romeo, August; Keil, Matthias

2010-05-19

Figure-ground is the segmentation of visual information into objects and their surrounding backgrounds. Two main processes herein are boundary assignment and surface segregation, which rely on the integration of global scene information. Recurrent processing either by intrinsic horizontal connections that connect surrounding neurons or by feedback projections from higher visual areas provide such information, and are considered to be the neural substrate for figure-ground segmentation. On the contrary, a role of feedforward projections in figure-ground segmentation is unknown. To have a better understanding of a role of feedforward connections in figure-ground organization, we constructed a feedforward spiking model using a biologically plausible neuron model. By means of surround inhibition our simple 3-layered model performs figure-ground segmentation and one-sided border-ownership coding. We propose that the visual system uses feed forward suppression for figure-ground segmentation and border-ownership assignment.

Feed-Forward Segmentation of Figure-Ground and Assignment of Border-Ownership

PubMed Central

Supèr, Hans; Romeo, August; Keil, Matthias

2010-01-01

Figure-ground is the segmentation of visual information into objects and their surrounding backgrounds. Two main processes herein are boundary assignment and surface segregation, which rely on the integration of global scene information. Recurrent processing either by intrinsic horizontal connections that connect surrounding neurons or by feedback projections from higher visual areas provide such information, and are considered to be the neural substrate for figure-ground segmentation. On the contrary, a role of feedforward projections in figure-ground segmentation is unknown. To have a better understanding of a role of feedforward connections in figure-ground organization, we constructed a feedforward spiking model using a biologically plausible neuron model. By means of surround inhibition our simple 3-layered model performs figure-ground segmentation and one-sided border-ownership coding. We propose that the visual system uses feed forward suppression for figure-ground segmentation and border-ownership assignment. PMID:20502718

Video Self-Modeling and Improving Oral Reading Fluency

ERIC Educational Resources Information Center

Chandler, Wanda Gail

2012-01-01

Self-modeling can take different forms but is described as a process where one observes one's own successful behavior and learns from it without dependence on any particular medium. In this study, two separate experiments were conducted to evaluate a video self-modeling (VSM) feedforward intervention. VSM feedforward (independent variable, IV),…

Artificial neural network modeling and optimization of ultrahigh pressure extraction of green tea polyphenols.

PubMed

Xi, Jun; Xue, Yujing; Xu, Yinxiang; Shen, Yuhong

2013-11-01

In this study, the ultrahigh pressure extraction of green tea polyphenols was modeled and optimized by a three-layer artificial neural network. A feed-forward neural network trained with an error back-propagation algorithm was used to evaluate the effects of pressure, liquid/solid ratio and ethanol concentration on the total phenolic content of green tea extracts. The neural network coupled with genetic algorithms was also used to optimize the conditions needed to obtain the highest yield of tea polyphenols. The obtained optimal architecture of artificial neural network model involved a feed-forward neural network with three input neurons, one hidden layer with eight neurons and one output layer including single neuron. The trained network gave the minimum value in the MSE of 0.03 and the maximum value in the R(2) of 0.9571, which implied a good agreement between the predicted value and the actual value, and confirmed a good generalization of the network. Based on the combination of neural network and genetic algorithms, the optimum extraction conditions for the highest yield of green tea polyphenols were determined as follows: 498.8 MPa for pressure, 20.8 mL/g for liquid/solid ratio and 53.6% for ethanol concentration. The total phenolic content of the actual measurement under the optimum predicated extraction conditions was 582.4 ± 0.63 mg/g DW, which was well matched with the predicted value (597.2mg/g DW). This suggests that the artificial neural network model described in this work is an efficient quantitative tool to predict the extraction efficiency of green tea polyphenols. Crown Copyright © 2013. Published by Elsevier Ltd. All rights reserved.

Robust high-performance control for robotic manipulators

NASA Technical Reports Server (NTRS)

Seraji, Homayoun (Inventor)

1991-01-01

Model-based and performance-based control techniques are combined for an electrical robotic control system. Thus, two distinct and separate design philosophies have been merged into a single control system having a control law formulation including two distinct and separate components, each of which yields a respective signal component that is combined into a total command signal for the system. Those two separate system components include a feedforward controller and a feedback controller. The feedforward controller is model-based and contains any known part of the manipulator dynamics that can be used for on-line control to produce a nominal feedforward component of the system's control signal. The feedback controller is performance-based and consists of a simple adaptive PID controller which generates an adaptive control signal to complement the nominal feedforward signal.

Robust high-performance control for robotic manipulators

NASA Technical Reports Server (NTRS)

Seraji, Homayoun (Inventor)

1989-01-01

Model-based and performance-based control techniques are combined for an electrical robotic control system. Thus, two distinct and separate design philosophies were merged into a single control system having a control law formulation including two distinct and separate components, each of which yields a respective signal componet that is combined into a total command signal for the system. Those two separate system components include a feedforward controller and feedback controller. The feedforward controller is model-based and contains any known part of the manipulator dynamics that can be used for on-line control to produce a nominal feedforward component of the system's control signal. The feedback controller is performance-based and consists of a simple adaptive PID controller which generates an adaptive control signal to complement the nomical feedforward signal.

Application of neural models as controllers in mobile robot velocity control loop

NASA Astrophysics Data System (ADS)

Cerkala, Jakub; Jadlovska, Anna

2017-01-01

This paper presents the application of an inverse neural models used as controllers in comparison to classical PI controllers for velocity tracking control task used in two-wheel, differentially driven mobile robot. The PI controller synthesis is based on linear approximation of actuators with equivalent load. In order to obtain relevant datasets for training of feed-forward multi-layer perceptron based neural network used as neural model, the mathematical model of mobile robot, that combines its kinematic and dynamic properties such as chassis dimensions, center of gravity offset, friction and actuator parameters is used. Neural models are trained off-line to act as an inverse dynamics of DC motors with particular load using data collected in simulation experiment for motor input voltage step changes within bounded operating area. The performances of PI controllers versus inverse neural models in mobile robot internal velocity control loops are demonstrated and compared in simulation experiment of navigation control task for line segment motion in plane.

Noise destroys feedback enhanced figure-ground segmentation but not feedforward figure-ground segmentation.

PubMed

Romeo, August; Arall, Marina; Supèr, Hans

2012-01-01

Figure-ground (FG) segmentation is the separation of visual information into background and foreground objects. In the visual cortex, FG responses are observed in the late stimulus response period, when neurons fire in tonic mode, and are accompanied by a switch in cortical state. When such a switch does not occur, FG segmentation fails. Currently, it is not known what happens in the brain on such occasions. A biologically plausible feedforward spiking neuron model was previously devised that performed FG segmentation successfully. After incorporating feedback the FG signal was enhanced, which was accompanied by a change in spiking regime. In a feedforward model neurons respond in a bursting mode whereas in the feedback model neurons fired in tonic mode. It is known that bursts can overcome noise, while tonic firing appears to be much more sensitive to noise. In the present study, we try to elucidate how the presence of noise can impair FG segmentation, and to what extent the feedforward and feedback pathways can overcome noise. We show that noise specifically destroys the feedback enhanced FG segmentation and leaves the feedforward FG segmentation largely intact. Our results predict that noise produces failure in FG perception.

Noise destroys feedback enhanced figure-ground segmentation but not feedforward figure-ground segmentation

PubMed Central

Romeo, August; Arall, Marina; Supèr, Hans

2012-01-01

Figure-ground (FG) segmentation is the separation of visual information into background and foreground objects. In the visual cortex, FG responses are observed in the late stimulus response period, when neurons fire in tonic mode, and are accompanied by a switch in cortical state. When such a switch does not occur, FG segmentation fails. Currently, it is not known what happens in the brain on such occasions. A biologically plausible feedforward spiking neuron model was previously devised that performed FG segmentation successfully. After incorporating feedback the FG signal was enhanced, which was accompanied by a change in spiking regime. In a feedforward model neurons respond in a bursting mode whereas in the feedback model neurons fired in tonic mode. It is known that bursts can overcome noise, while tonic firing appears to be much more sensitive to noise. In the present study, we try to elucidate how the presence of noise can impair FG segmentation, and to what extent the feedforward and feedback pathways can overcome noise. We show that noise specifically destroys the feedback enhanced FG segmentation and leaves the feedforward FG segmentation largely intact. Our results predict that noise produces failure in FG perception. PMID:22934028

Can Video Self-Modeling Improve Affected Limb Reach and Grasp Ability in Stroke Patients?

PubMed

Steel, Kylie Ann; Mudie, Kurt; Sandoval, Remi; Anderson, David; Dogramaci, Sera; Rehmanjan, Mohammad; Birznieks, Ingvars

2018-01-01

The authors examined whether feedforward video self-modeling (FF VSM) would improve control over the affected limb, movement self-confidence, movement self-consciousness, and well-being in 18 stroke survivors. Participants completed a cup transport task and 2 questionnaires related to psychological processes pre- and postintervention. Pretest video footage of the unaffected limb performing the task was edited to create a best-of or mirror-reversed training DVD, creating the illusion that patients were performing proficiently with the affected limb. The training yielded significant improvements for the forward movement of the affected limb compared to the unaffected limb. Significant improvements were also seen in movement self-confidence, movement self-consciousness, and well-being. FF VSM appears to be a viable way to improve motor ability in populations with movement disorders.

An on-line modified least-mean-square algorithm for training neurofuzzy controllers.

PubMed

Tan, Woei Wan

2007-04-01

The problem hindering the use of data-driven modelling methods for training controllers on-line is the lack of control over the amount by which the plant is excited. As the operating schedule determines the information available on-line, the knowledge of the process may degrade if the setpoint remains constant for an extended period. This paper proposes an identification algorithm that alleviates "learning interference" by incorporating fuzzy theory into the normalized least-mean-square update rule. The ability of the proposed methodology to achieve faster learning is examined by employing the algorithm to train a neurofuzzy feedforward controller for controlling a liquid level process. Since the proposed identification strategy has similarities with the normalized least-mean-square update rule and the recursive least-square estimator, the on-line learning rates of these algorithms are also compared.

Optimization of Artificial Neural Network using Evolutionary Programming for Prediction of Cascading Collapse Occurrence due to the Hidden Failure Effect

NASA Astrophysics Data System (ADS)

Idris, N. H.; Salim, N. A.; Othman, M. M.; Yasin, Z. M.

2018-03-01

This paper presents the Evolutionary Programming (EP) which proposed to optimize the training parameters for Artificial Neural Network (ANN) in predicting cascading collapse occurrence due to the effect of protection system hidden failure. The data has been collected from the probability of hidden failure model simulation from the historical data. The training parameters of multilayer-feedforward with backpropagation has been optimized with objective function to minimize the Mean Square Error (MSE). The optimal training parameters consists of the momentum rate, learning rate and number of neurons in first hidden layer and second hidden layer is selected in EP-ANN. The IEEE 14 bus system has been tested as a case study to validate the propose technique. The results show the reliable prediction of performance validated through MSE and Correlation Coefficient (R).

Numerical solution of the nonlinear Schrodinger equation by feedforward neural networks

NASA Astrophysics Data System (ADS)

Shirvany, Yazdan; Hayati, Mohsen; Moradian, Rostam

2008-12-01

We present a method to solve boundary value problems using artificial neural networks (ANN). A trial solution of the differential equation is written as a feed-forward neural network containing adjustable parameters (the weights and biases). From the differential equation and its boundary conditions we prepare the energy function which is used in the back-propagation method with momentum term to update the network parameters. We improved energy function of ANN which is derived from Schrodinger equation and the boundary conditions. With this improvement of energy function we can use unsupervised training method in the ANN for solving the equation. Unsupervised training aims to minimize a non-negative energy function. We used the ANN method to solve Schrodinger equation for few quantum systems. Eigenfunctions and energy eigenvalues are calculated. Our numerical results are in agreement with their corresponding analytical solution and show the efficiency of ANN method for solving eigenvalue problems.

Stimulus-specific adaptation in a recurrent network model of primary auditory cortex

PubMed Central

2017-01-01

Stimulus-specific adaptation (SSA) occurs when neurons decrease their responses to frequently-presented (standard) stimuli but not, or not as much, to other, rare (deviant) stimuli. SSA is present in all mammalian species in which it has been tested as well as in birds. SSA confers short-term memory to neuronal responses, and may lie upstream of the generation of mismatch negativity (MMN), an important human event-related potential. Previously published models of SSA mostly rely on synaptic depression of the feedforward, thalamocortical input. Here we study SSA in a recurrent neural network model of primary auditory cortex. When the recurrent, intracortical synapses display synaptic depression, the network generates population spikes (PSs). SSA occurs in this network when deviants elicit a PS but standards do not, and we demarcate the regions in parameter space that allow SSA. While SSA based on PSs does not require feedforward depression, we identify feedforward depression as a mechanism for expanding the range of parameters that support SSA. We provide predictions for experiments that could help differentiate between SSA due to synaptic depression of feedforward connections and SSA due to synaptic depression of recurrent connections. Similar to experimental data, the magnitude of SSA in the model depends on the frequency difference between deviant and standard, probability of the deviant, inter-stimulus interval and input amplitude. In contrast to models based on feedforward depression, our model shows true deviance sensitivity as found in experiments. PMID:28288158

Noise Tolerance of Attractor and Feedforward Memory Models

PubMed Central

Lim, Sukbin; Goldman, Mark S.

2017-01-01

In short-term memory networks, transient stimuli are represented by patterns of neural activity that persist long after stimulus offset. Here, we compare the performance of two prominent classes of memory networks, feedback-based attractor networks and feedforward networks, in conveying information about the amplitude of a briefly presented stimulus in the presence of gaussian noise. Using Fisher information as a metric of memory performance, we find that the optimal form of network architecture depends strongly on assumptions about the forms of nonlinearities in the network. For purely linear networks, we find that feedforward networks outperform attractor networks because noise is continually removed from feedforward networks when signals exit the network; as a result, feedforward networks can amplify signals they receive faster than noise accumulates over time. By contrast, attractor networks must operate in a signal-attenuating regime to avoid the buildup of noise. However, if the amplification of signals is limited by a finite dynamic range of neuronal responses or if noise is reset at the time of signal arrival, as suggested by recent experiments, we find that attractor networks can out-perform feedforward ones. Under a simple model in which neurons have a finite dynamic range, we find that the optimal attractor networks are forgetful if there is no mechanism for noise reduction with signal arrival but nonforgetful (perfect integrators) in the presence of a strong reset mechanism. Furthermore, we find that the maximal Fisher information for the feedforward and attractor networks exhibits power law decay as a function of time and scales linearly with the number of neurons. These results highlight prominent factors that lead to trade-offs in the memory performance of networks with different architectures and constraints, and suggest conditions under which attractor or feedforward networks may be best suited to storing information about previous stimuli. PMID:22091664

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising.

PubMed

Zhang, Kai; Zuo, Wangmeng; Chen, Yunjin; Meng, Deyu; Zhang, Lei

2017-07-01

The discriminative model learning for image denoising has been recently attracting considerable attentions due to its favorable denoising performance. In this paper, we take one step forward by investigating the construction of feed-forward denoising convolutional neural networks (DnCNNs) to embrace the progress in very deep architecture, learning algorithm, and regularization method into image denoising. Specifically, residual learning and batch normalization are utilized to speed up the training process as well as boost the denoising performance. Different from the existing discriminative denoising models which usually train a specific model for additive white Gaussian noise at a certain noise level, our DnCNN model is able to handle Gaussian denoising with unknown noise level (i.e., blind Gaussian denoising). With the residual learning strategy, DnCNN implicitly removes the latent clean image in the hidden layers. This property motivates us to train a single DnCNN model to tackle with several general image denoising tasks, such as Gaussian denoising, single image super-resolution, and JPEG image deblocking. Our extensive experiments demonstrate that our DnCNN model can not only exhibit high effectiveness in several general image denoising tasks, but also be efficiently implemented by benefiting from GPU computing.

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.

Model reference adaptive control of flexible robots in the presence of sudden load changes

NASA Technical Reports Server (NTRS)

Steinvorth, Rodrigo; Kaufman, Howard; Neat, Gregory

1991-01-01

Direct command generator tracker based model reference adaptive control (MRAC) algorithms are applied to the dynamics for a flexible-joint arm in the presence of sudden load changes. Because of the need to satisfy a positive real condition, such MRAC procedures are designed so that a feedforward augmented output follows the reference model output, thus, resulting in an ultimately bounded rather than zero output error. Thus, modifications are suggested and tested that: (1) incorporate feedforward into the reference model's output as well as the plant's output, and (2) incorporate a derivative term into only the process feedforward loop. The results of these simulations give a response with zero steady state model following error, and thus encourage further use of MRAC for more complex flexibile robotic systems.

Lateral and feedforward inhibition suppress asynchronous activity in a large, biophysically-detailed computational model of the striatal network

PubMed Central

Moyer, Jason T.; Halterman, Benjamin L.; Finkel, Leif H.; Wolf, John A.

2014-01-01

Striatal medium spiny neurons (MSNs) receive lateral inhibitory projections from other MSNs and feedforward inhibitory projections from fast-spiking, parvalbumin-containing striatal interneurons (FSIs). The functional roles of these connections are unknown, and difficult to study in an experimental preparation. We therefore investigated the functionality of both lateral (MSN-MSN) and feedforward (FSI-MSN) inhibition using a large-scale computational model of the striatal network. The model consists of 2744 MSNs comprised of 189 compartments each and 121 FSIs comprised of 148 compartments each, with dendrites explicitly represented and almost all known ionic currents included and strictly constrained by biological data as appropriate. Our analysis of the model indicates that both lateral inhibition and feedforward inhibition function at the population level to limit non-ensemble MSN spiking while preserving ensemble MSN spiking. Specifically, lateral inhibition enables large ensembles of MSNs firing synchronously to strongly suppress non-ensemble MSNs over a short time-scale (10–30 ms). Feedforward inhibition enables FSIs to strongly inhibit weakly activated, non-ensemble MSNs while moderately inhibiting activated ensemble MSNs. Importantly, FSIs appear to more effectively inhibit MSNs when FSIs fire asynchronously. Both types of inhibition would increase the signal-to-noise ratio of responding MSN ensembles and contribute to the formation and dissolution of MSN ensembles in the striatal network. PMID:25505406

Using an Extended Kalman Filter Learning Algorithm for Feed-Forward Neural Networks to Describe Tracer Correlations

NASA Technical Reports Server (NTRS)

Lary, David J.; Mussa, Yussuf

2004-01-01

In this study a new extended Kalman filter (EKF) learning algorithm for feed-forward neural networks (FFN) is used. With the EKF approach, the training of the FFN can be seen as state estimation for a non-linear stationary process. The EKF method gives excellent convergence performances provided that there is enough computer core memory and that the machine precision is high. Neural networks are ideally suited to describe the spatial and temporal dependence of tracer-tracer correlations. The neural network performs well even in regions where the correlations are less compact and normally a family of correlation curves would be required. For example, the CH4-N2O correlation can be well described using a neural network trained with the latitude, pressure, time of year, and CH4 volume mixing ratio (v.m.r.). The neural network was able to reproduce the CH4-N2O correlation with a correlation coefficient between simulated and training values of 0.9997. The neural network Fortran code used is available for download.

Hebbian learning in a model with dynamic rate-coded neurons: an alternative to the generative model approach for learning receptive fields from natural scenes.

PubMed

Hamker, Fred H; Wiltschut, Jan

2007-09-01

Most computational models of coding are based on a generative model according to which the feedback signal aims to reconstruct the visual scene as close as possible. We here explore an alternative model of feedback. It is derived from studies of attention and thus, probably more flexible with respect to attentive processing in higher brain areas. According to this model, feedback implements a gain increase of the feedforward signal. We use a dynamic model with presynaptic inhibition and Hebbian learning to simultaneously learn feedforward and feedback weights. The weights converge to localized, oriented, and bandpass filters similar as the ones found in V1. Due to presynaptic inhibition the model predicts the organization of receptive fields within the feedforward pathway, whereas feedback primarily serves to tune early visual processing according to the needs of the task.

Learning polynomial feedforward neural networks by genetic programming and backpropagation.

PubMed

Nikolaev, N Y; Iba, H

2003-01-01

This paper presents an approach to learning polynomial feedforward neural networks (PFNNs). The approach suggests, first, finding the polynomial network structure by means of a population-based search technique relying on the genetic programming paradigm, and second, further adjustment of the best discovered network weights by an especially derived backpropagation algorithm for higher order networks with polynomial activation functions. These two stages of the PFNN learning process enable us to identify networks with good training as well as generalization performance. Empirical results show that this approach finds PFNN which outperform considerably some previous constructive polynomial network algorithms on processing benchmark time series.

Quantum generalisation of feedforward neural networks

NASA Astrophysics Data System (ADS)

Wan, Kwok Ho; Dahlsten, Oscar; Kristjánsson, Hlér; Gardner, Robert; Kim, M. S.

2017-09-01

We propose a quantum generalisation of a classical neural network. The classical neurons are firstly rendered reversible by adding ancillary bits. Then they are generalised to being quantum reversible, i.e., unitary (the classical networks we generalise are called feedforward, and have step-function activation functions). The quantum network can be trained efficiently using gradient descent on a cost function to perform quantum generalisations of classical tasks. We demonstrate numerically that it can: (i) compress quantum states onto a minimal number of qubits, creating a quantum autoencoder, and (ii) discover quantum communication protocols such as teleportation. Our general recipe is theoretical and implementation-independent. The quantum neuron module can naturally be implemented photonically.

Predictive Feedback and Feedforward Control for Systems with Unknown Disturbances

NASA Technical Reports Server (NTRS)

Juang, Jer-Nan; Eure, Kenneth W.

1998-01-01

Predictive feedback control has been successfully used in the regulation of plate vibrations when no reference signal is available for feedforward control. However, if a reference signal is available it may be used to enhance regulation by incorporating a feedforward path in the feedback controller. Such a controller is known as a hybrid controller. This paper presents the theory and implementation of the hybrid controller for general linear systems, in particular for structural vibration induced by acoustic noise. The generalized predictive control is extended to include a feedforward path in the multi-input multi-output case and implemented on a single-input single-output test plant to achieve plate vibration regulation. There are cases in acoustic-induce vibration where the disturbance signal is not available to be used by the hybrid controller, but a disturbance model is available. In this case the disturbance model may be used in the feedback controller to enhance performance. In practice, however, neither the disturbance signal nor the disturbance model is available. This paper presents the theory of identifying and incorporating the noise model into the feedback controller. Implementations are performed on a test plant and regulation improvements over the case where no noise model is used are demonstrated.

Character Recognition Using Genetically Trained Neural Networks

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

Diniz, C.; Stantz, K.M.; Trahan, M.W.

1998-10-01

Computationally intelligent recognition of characters and symbols addresses a wide range of applications including foreign language translation and chemical formula identification. The combination of intelligent learning and optimization algorithms with layered neural structures offers powerful techniques for character recognition. These techniques were originally developed by Sandia National Laboratories for pattern and spectral analysis; however, their ability to optimize vast amounts of data make them ideal for character recognition. An adaptation of the Neural Network Designer soflsvare allows the user to create a neural network (NN_) trained by a genetic algorithm (GA) that correctly identifies multiple distinct characters. The initial successfidmore » recognition of standard capital letters can be expanded to include chemical and mathematical symbols and alphabets of foreign languages, especially Arabic and Chinese. The FIN model constructed for this project uses a three layer feed-forward architecture. To facilitate the input of characters and symbols, a graphic user interface (GUI) has been developed to convert the traditional representation of each character or symbol to a bitmap. The 8 x 8 bitmap representations used for these tests are mapped onto the input nodes of the feed-forward neural network (FFNN) in a one-to-one correspondence. The input nodes feed forward into a hidden layer, and the hidden layer feeds into five output nodes correlated to possible character outcomes. During the training period the GA optimizes the weights of the NN until it can successfully recognize distinct characters. Systematic deviations from the base design test the network's range of applicability. Increasing capacity, the number of letters to be recognized, requires a nonlinear increase in the number of hidden layer neurodes. Optimal character recognition performance necessitates a minimum threshold for the number of cases when genetically training the net. And, the amount of noise significantly degrades character recognition efficiency, some of which can be overcome by adding noise during training and optimizing the form of the network's activation fimction.« less

Effects of topologies on signal propagation in feedforward networks

NASA Astrophysics Data System (ADS)

Zhao, Jia; Qin, Ying-Mei; Che, Yan-Qiu

2018-01-01

We systematically investigate the effects of topologies on signal propagation in feedforward networks (FFNs) based on the FitzHugh-Nagumo neuron model. FFNs with different topological structures are constructed with same number of both in-degrees and out-degrees in each layer and given the same input signal. The propagation of firing patterns and firing rates are found to be affected by the distribution of neuron connections in the FFNs. Synchronous firing patterns emerge in the later layers of FFNs with identical, uniform, and exponential degree distributions, but the number of synchronous spike trains in the output layers of the three topologies obviously differs from one another. The firing rates in the output layers of the three FFNs can be ordered from high to low according to their topological structures as exponential, uniform, and identical distributions, respectively. Interestingly, the sequence of spiking regularity in the output layers of the three FFNs is consistent with the firing rates, but their firing synchronization is in the opposite order. In summary, the node degree is an important factor that can dramatically influence the neuronal network activity.

A dynamic feedforward neural network based on gaussian particle swarm optimization and its application for predictive control.

PubMed

Han, Min; Fan, Jianchao; Wang, Jun

2011-09-01

A dynamic feedforward neural network (DFNN) is proposed for predictive control, whose adaptive parameters are adjusted by using Gaussian particle swarm optimization (GPSO) in the training process. Adaptive time-delay operators are added in the DFNN to improve its generalization for poorly known nonlinear dynamic systems with long time delays. Furthermore, GPSO adopts a chaotic map with Gaussian function to balance the exploration and exploitation capabilities of particles, which improves the computational efficiency without compromising the performance of the DFNN. The stability of the particle dynamics is analyzed, based on the robust stability theory, without any restrictive assumption. A stability condition for the GPSO+DFNN model is derived, which ensures a satisfactory global search and quick convergence, without the need for gradients. The particle velocity ranges could change adaptively during the optimization process. The results of a comparative study show that the performance of the proposed algorithm can compete with selected algorithms on benchmark problems. Additional simulation results demonstrate the effectiveness and accuracy of the proposed combination algorithm in identifying and controlling nonlinear systems with long time delays.

Effects of topologies on signal propagation in feedforward networks.

PubMed

Zhao, Jia; Qin, Ying-Mei; Che, Yan-Qiu

2018-01-01

We systematically investigate the effects of topologies on signal propagation in feedforward networks (FFNs) based on the FitzHugh-Nagumo neuron model. FFNs with different topological structures are constructed with same number of both in-degrees and out-degrees in each layer and given the same input signal. The propagation of firing patterns and firing rates are found to be affected by the distribution of neuron connections in the FFNs. Synchronous firing patterns emerge in the later layers of FFNs with identical, uniform, and exponential degree distributions, but the number of synchronous spike trains in the output layers of the three topologies obviously differs from one another. The firing rates in the output layers of the three FFNs can be ordered from high to low according to their topological structures as exponential, uniform, and identical distributions, respectively. Interestingly, the sequence of spiking regularity in the output layers of the three FFNs is consistent with the firing rates, but their firing synchronization is in the opposite order. In summary, the node degree is an important factor that can dramatically influence the neuronal network activity.

Visual crowding illustrates the inadequacy of local vs. global and feedforward vs. feedback distinctions in modeling visual perception

PubMed Central

Clarke, Aaron M.; Herzog, Michael H.; Francis, Gregory

2014-01-01

Experimentalists tend to classify models of visual perception as being either local or global, and involving either feedforward or feedback processing. We argue that these distinctions are not as helpful as they might appear, and we illustrate these issues by analyzing models of visual crowding as an example. Recent studies have argued that crowding cannot be explained by purely local processing, but that instead, global factors such as perceptual grouping are crucial. Theories of perceptual grouping, in turn, often invoke feedback connections as a way to account for their global properties. We examined three types of crowding models that are representative of global processing models, and two of which employ feedback processing: a model based on Fourier filtering, a feedback neural network, and a specific feedback neural architecture that explicitly models perceptual grouping. Simulations demonstrate that crucial empirical findings are not accounted for by any of the models. We conclude that empirical investigations that reject a local or feedforward architecture offer almost no constraints for model construction, as there are an uncountable number of global and feedback systems. We propose that the identification of a system as being local or global and feedforward or feedback is less important than the identification of a system's computational details. Only the latter information can provide constraints on model development and promote quantitative explanations of complex phenomena. PMID:25374554

Feed-forward transcriptional programming by nuclear receptors: regulatory principles and therapeutic implications.

PubMed

Sasse, Sarah K; Gerber, Anthony N

2015-01-01

Nuclear receptors (NRs) are widely targeted to treat a range of human diseases. Feed-forward loops are an ancient mechanism through which single cell organisms organize transcriptional programming and modulate gene expression dynamics, but they have not been systematically studied as a regulatory paradigm for NR-mediated transcriptional responses. Here, we provide an overview of the basic properties of feed-forward loops as predicted by mathematical models and validated experimentally in single cell organisms. We review existing evidence implicating feed-forward loops as important in controlling clinically relevant transcriptional responses to estrogens, progestins, and glucocorticoids, among other NR ligands. We propose that feed-forward transcriptional circuits are a major mechanism through which NRs integrate signals, exert temporal control over gene regulation, and compartmentalize client transcriptomes into discrete subunits. Implications for the design and function of novel selective NR ligands are discussed. Copyright © 2014 Elsevier Inc. All rights reserved.

Training trajectories by continuous recurrent multilayer networks.

PubMed

Leistritz, L; Galicki, M; Witte, H; Kochs, E

2002-01-01

This paper addresses the problem of training trajectories by means of continuous recurrent neural networks whose feedforward parts are multilayer perceptrons. Such networks can approximate a general nonlinear dynamic system with arbitrary accuracy. The learning process is transformed into an optimal control framework where the weights are the controls to be determined. A training algorithm based upon a variational formulation of Pontryagin's maximum principle is proposed for such networks. Computer examples demonstrating the efficiency of the given approach are also presented.

Stochastic resonance in feedforward acupuncture networks

NASA Astrophysics Data System (ADS)

Qin, Ying-Mei; Wang, Jiang; Men, Cong; Deng, Bin; Wei, Xi-Le; Yu, Hai-Tao; Chan, Wai-Lok

2014-10-01

Effects of noises and some other network properties on the weak signal propagation are studied systematically in feedforward acupuncture networks (FFN) based on FitzHugh-Nagumo neuron model. It is found that noises with medium intensity can enhance signal propagation and this effect can be further increased by the feedforward network structure. Resonant properties in the noisy network can also be altered by several network parameters, such as heterogeneity, synapse features, and feedback connections. These results may also provide a novel potential explanation for the propagation of acupuncture signal.

Neural dynamics of feedforward and feedback processing in figure-ground segregation

PubMed Central

Layton, Oliver W.; Mingolla, Ennio; Yazdanbakhsh, Arash

2014-01-01

Determining whether a region belongs to the interior or exterior of a shape (figure-ground segregation) is a core competency of the primate brain, yet the underlying mechanisms are not well understood. Many models assume that figure-ground segregation occurs by assembling progressively more complex representations through feedforward connections, with feedback playing only a modulatory role. We present a dynamical model of figure-ground segregation in the primate ventral stream wherein feedback plays a crucial role in disambiguating a figure's interior and exterior. We introduce a processing strategy whereby jitter in RF center locations and variation in RF sizes is exploited to enhance and suppress neural activity inside and outside of figures, respectively. Feedforward projections emanate from units that model cells in V4 known to respond to the curvature of boundary contours (curved contour cells), and feedback projections from units predicted to exist in IT that strategically group neurons with different RF sizes and RF center locations (teardrop cells). Neurons (convex cells) that preferentially respond when centered on a figure dynamically balance feedforward (bottom-up) information and feedback from higher visual areas. The activation is enhanced when an interior portion of a figure is in the RF via feedback from units that detect closure in the boundary contours of a figure. Our model produces maximal activity along the medial axis of well-known figures with and without concavities, and inside algorithmically generated shapes. Our results suggest that the dynamic balancing of feedforward signals with the specific feedback mechanisms proposed by the model is crucial for figure-ground segregation. PMID:25346703

Neural dynamics of feedforward and feedback processing in figure-ground segregation.

PubMed

Layton, Oliver W; Mingolla, Ennio; Yazdanbakhsh, Arash

2014-01-01

Determining whether a region belongs to the interior or exterior of a shape (figure-ground segregation) is a core competency of the primate brain, yet the underlying mechanisms are not well understood. Many models assume that figure-ground segregation occurs by assembling progressively more complex representations through feedforward connections, with feedback playing only a modulatory role. We present a dynamical model of figure-ground segregation in the primate ventral stream wherein feedback plays a crucial role in disambiguating a figure's interior and exterior. We introduce a processing strategy whereby jitter in RF center locations and variation in RF sizes is exploited to enhance and suppress neural activity inside and outside of figures, respectively. Feedforward projections emanate from units that model cells in V4 known to respond to the curvature of boundary contours (curved contour cells), and feedback projections from units predicted to exist in IT that strategically group neurons with different RF sizes and RF center locations (teardrop cells). Neurons (convex cells) that preferentially respond when centered on a figure dynamically balance feedforward (bottom-up) information and feedback from higher visual areas. The activation is enhanced when an interior portion of a figure is in the RF via feedback from units that detect closure in the boundary contours of a figure. Our model produces maximal activity along the medial axis of well-known figures with and without concavities, and inside algorithmically generated shapes. Our results suggest that the dynamic balancing of feedforward signals with the specific feedback mechanisms proposed by the model is crucial for figure-ground segregation.

Performance Assessment of Model-Based Optimal Feedforward and Feedback Current Profile Control in NSTX-U using the TRANSP Code

NASA Astrophysics Data System (ADS)

Ilhan, Z.; Wehner, W. P.; Schuster, E.; Boyer, M. D.; Gates, D. A.; Gerhardt, S.; Menard, J.

2015-11-01

Active control of the toroidal current density profile is crucial to achieve and maintain high-performance, MHD-stable plasma operation in NSTX-U. A first-principles-driven, control-oriented model describing the temporal evolution of the current profile has been proposed earlier by combining the magnetic diffusion equation with empirical correlations obtained at NSTX-U for the electron density, electron temperature, and non-inductive current drives. A feedforward + feedback control scheme for the requlation of the current profile is constructed by embedding the proposed nonlinear, physics-based model into the control design process. Firstly, nonlinear optimization techniques are used to design feedforward actuator trajectories that steer the plasma to a desired operating state with the objective of supporting the traditional trial-and-error experimental process of advanced scenario planning. Secondly, a feedback control algorithm to track a desired current profile evolution is developed with the goal of adding robustness to the overall control scheme. The effectiveness of the combined feedforward + feedback control algorithm for current profile regulation is tested in predictive simulations carried out in TRANSP. Supported by PPPL.

Sliding-mode control combined with improved adaptive feedforward for wafer scanner

NASA Astrophysics Data System (ADS)

Li, Xiaojie; Wang, Yiguang

2018-03-01

In this paper, a sliding-mode control method combined with improved adaptive feedforward is proposed for wafer scanner to improve the tracking performance of the closed-loop system. Particularly, In addition to the inverse model, the nonlinear force ripple effect which may degrade the tracking accuracy of permanent magnet linear motor (PMLM) is considered in the proposed method. The dominant position periodicity of force ripple is determined by using the Fast Fourier Transform (FFT) analysis for experimental data and the improved feedforward control is achieved by the online recursive least-squares (RLS) estimation of the inverse model and the force ripple. The improved adaptive feedforward is given in a general form of nth-order model with force ripple effect. This proposed method is motivated by the motion controller design of the long-stroke PMLM and short-stroke voice coil motor for wafer scanner. The stability of the closed-loop control system and the convergence of the motion tracking are guaranteed by the proposed sliding-mode feedback and adaptive feedforward methods theoretically. Comparative experiments on a precision linear motion platform can verify the correctness and effectiveness of the proposed method. The experimental results show that comparing to traditional method the proposed one has better performance of rapidity and robustness, especially for high speed motion trajectory. And, the improvements on both tracking accuracy and settling time can be achieved.

Adaptive Neuron Model: An architecture for the rapid learning of nonlinear topological transformations

NASA Technical Reports Server (NTRS)

Tawel, Raoul (Inventor)

1994-01-01

A method for the rapid learning of nonlinear mappings and topological transformations using a dynamically reconfigurable artificial neural network is presented. This fully-recurrent Adaptive Neuron Model (ANM) network was applied to the highly degenerate inverse kinematics problem in robotics, and its performance evaluation is bench-marked. Once trained, the resulting neuromorphic architecture was implemented in custom analog neural network hardware and the parameters capturing the functional transformation downloaded onto the system. This neuroprocessor, capable of 10(exp 9) ops/sec, was interfaced directly to a three degree of freedom Heathkit robotic manipulator. Calculation of the hardware feed-forward pass for this mapping was benchmarked at approximately 10 microsec.

The modelling of lead removal from water by deep eutectic solvents functionalized CNTs: artificial neural network (ANN) approach.

PubMed

Fiyadh, Seef Saadi; AlSaadi, Mohammed Abdulhakim; AlOmar, Mohamed Khalid; Fayaed, Sabah Saadi; Hama, Ako R; Bee, Sharifah; El-Shafie, Ahmed

2017-11-01

The main challenge in the lead removal simulation is the behaviour of non-linearity relationships between the process parameters. The conventional modelling technique usually deals with this problem by a linear method. The substitute modelling technique is an artificial neural network (ANN) system, and it is selected to reflect the non-linearity in the interaction among the variables in the function. Herein, synthesized deep eutectic solvents were used as a functionalized agent with carbon nanotubes as adsorbents of Pb 2+ . Different parameters were used in the adsorption study including pH (2.7 to 7), adsorbent dosage (5 to 20 mg), contact time (3 to 900 min) and Pb 2+ initial concentration (3 to 60 mg/l). The number of experimental trials to feed and train the system was 158 runs conveyed in laboratory scale. Two ANN types were designed in this work, the feed-forward back-propagation and layer recurrent; both methods are compared based on their predictive proficiency in terms of the mean square error (MSE), root mean square error, relative root mean square error, mean absolute percentage error and determination coefficient (R 2 ) based on the testing dataset. The ANN model of lead removal was subjected to accuracy determination and the results showed R 2 of 0.9956 with MSE of 1.66 × 10 -4 . The maximum relative error is 14.93% for the feed-forward back-propagation neural network model.

Perceptual Learning via Modification of Cortical Top-Down Signals

PubMed Central

Schäfer, Roland; Vasilaki, Eleni; Senn, Walter

2007-01-01

The primary visual cortex (V1) is pre-wired to facilitate the extraction of behaviorally important visual features. Collinear edge detectors in V1, for instance, mutually enhance each other to improve the perception of lines against a noisy background. The same pre-wiring that facilitates line extraction, however, is detrimental when subjects have to discriminate the brightness of different line segments. How is it possible to improve in one task by unsupervised practicing, without getting worse in the other task? The classical view of perceptual learning is that practicing modulates the feedforward input stream through synaptic modifications onto or within V1. However, any rewiring of V1 would deteriorate other perceptual abilities different from the trained one. We propose a general neuronal model showing that perceptual learning can modulate top-down input to V1 in a task-specific way while feedforward and lateral pathways remain intact. Consistent with biological data, the model explains how context-dependent brightness discrimination is improved by a top-down recruitment of recurrent inhibition and a top-down induced increase of the neuronal gain within V1. Both the top-down modulation of inhibition and of neuronal gain are suggested to be universal features of cortical microcircuits which enable perceptual learning. PMID:17715996

A computational feedforward model predicts categorization of masked emotional body language for longer, but not for shorter, latencies.

PubMed

Stienen, Bernard M C; Schindler, Konrad; de Gelder, Beatrice

2012-07-01

Given the presence of massive feedback loops in brain networks, it is difficult to disentangle the contribution of feedforward and feedback processing to the recognition of visual stimuli, in this case, of emotional body expressions. The aim of the work presented in this letter is to shed light on how well feedforward processing explains rapid categorization of this important class of stimuli. By means of parametric masking, it may be possible to control the contribution of feedback activity in human participants. A close comparison is presented between human recognition performance and the performance of a computational neural model that exclusively modeled feedforward processing and was engineered to fulfill the computational requirements of recognition. Results show that the longer the stimulus onset asynchrony (SOA), the closer the performance of the human participants was to the values predicted by the model, with an optimum at an SOA of 100 ms. At short SOA latencies, human performance deteriorated, but the categorization of the emotional expressions was still above baseline. The data suggest that, although theoretically, feedback arising from inferotemporal cortex is likely to be blocked when the SOA is 100 ms, human participants still seem to rely on more local visual feedback processing to equal the model's performance.

Stability analysis of feedforward anticipation optimal flux difference in traffic lattice hydrodynamic theory

NASA Astrophysics Data System (ADS)

Sun, Di-Hua; Zhang, Geng; Zhao, Min; Cheng, Sen-Lin; Cao, Jian-Dong

2018-03-01

Recently, the influence of driver's individual behaviors on traffic stability is research hotspot with the fasting developing transportation cyber-physical systems. In this paper, a new traffic lattice hydrodynamic model is proposed with consideration of driver's feedforward anticipation optimal flux difference. The neutral stability condition of the new model is obtained through linear stability analysis theory. The results show that the stable region will be enlarged on the phase diagram when the feedforward anticipation optimal flux difference effect is taken into account. In order to depict traffic jamming transition properties theoretically, the mKdV equation near the critical point is derived via nonlinear reductive perturbation method. The propagation behavior of traffic density waves can be described by the kink-antikink solution of the mKdV equation. Numerical simulations are conducted to verify the analytical results and all the results confirms that traffic stability can be enhanced significantly by considering the feedforward anticipation optimal flux difference in traffic lattice hydrodynamic theory.

[Feedforward control strategy and its application in quality improvement of ethanol precipitation process of danhong injection].

PubMed

Yan, Bin-Jun; Guo, Zheng-Tai; Qu, Hai-Bin; Zhao, Bu-Chang; Zhao, Tao

2013-06-01

In this work, a feedforward control strategy basing on the concept of quality by design was established for the manufacturing process of traditional Chinese medicine to reduce the impact of the quality variation of raw materials on drug. In the research, the ethanol precipitation process of Danhong injection was taken as an application case of the method established. Box-Behnken design of experiments was conducted. Mathematical models relating the attributes of the concentrate, the process parameters and the quality of the supernatants produced were established. Then an optimization model for calculating the best process parameters basing on the attributes of the concentrate was built. The quality of the supernatants produced by ethanol precipitation with optimized and non-optimized process parameters were compared. The results showed that using the feedforward control strategy for process parameters optimization can control the quality of the supernatants effectively. The feedforward control strategy proposed can enhance the batch-to-batch consistency of the supernatants produced by ethanol precipitation.

Development of experimental design approach and ANN-based models for determination of Cr(VI) ions uptake rate from aqueous solution onto the solid biodiesel waste residue.

PubMed

Shanmugaprakash, M; Sivakumar, V

2013-11-01

In the present work, the evaluation capacities of two optimization methodologies such as RSM and ANN were employed and compared for predication of Cr(VI) uptake rate using defatted pongamia oil cake (DPOC) in both batch and column mode. The influence of operating parameters was investigated through a central composite design (CCD) of RSM using Design Expert 8.0.7.1 software. The same data was fed as input in ANN to obtain a trained the multilayer feed-forward networks back-propagation algorithm using MATLAB. The performance of the developed ANN models were compared with RSM mathematical models for Cr(VI) uptake rate in terms of the coefficient of determination (R(2)), root mean square error (RMSE) and absolute average deviation (AAD). The estimated values confirm that ANN predominates RSM representing the superiority of a trained ANN models over RSM models in order to capture the non-linear behavior of the given system. Copyright © 2013 Elsevier Ltd. All rights reserved.

Predicting high-risk preterm birth using artificial neural networks.

PubMed

Catley, Christina; Frize, Monique; Walker, C Robin; Petriu, Dorina C

2006-07-01

A reengineered approach to the early prediction of preterm birth is presented as a complimentary technique to the current procedure of using costly and invasive clinical testing on high-risk maternal populations. Artificial neural networks (ANNs) are employed as a screening tool for preterm birth on a heterogeneous maternal population; risk estimations use obstetrical variables available to physicians before 23 weeks gestation. The objective was to assess if ANNs have a potential use in obstetrical outcome estimations in low-risk maternal populations. The back-propagation feedforward ANN was trained and tested on cases with eight input variables describing the patient's obstetrical history; the output variables were: 1) preterm birth; 2) high-risk preterm birth; and 3) a refined high-risk preterm birth outcome excluding all cases where resuscitation was delivered in the form of free flow oxygen. Artificial training sets were created to increase the distribution of the underrepresented class to 20%. Training on the refined high-risk preterm birth model increased the network's sensitivity to 54.8%, compared to just over 20% for the nonartificially distributed preterm birth model.

Role of intraglomerular circuits in shaping temporally structured responses to naturalistic inhalation-driven sensory input to the olfactory bulb

PubMed Central

Carey, Ryan M.; Sherwood, William Erik; Shipley, Michael T.; Borisyuk, Alla

2015-01-01

Olfaction in mammals is a dynamic process driven by the inhalation of air through the nasal cavity. Inhalation determines the temporal structure of sensory neuron responses and shapes the neural dynamics underlying central olfactory processing. Inhalation-linked bursts of activity among olfactory bulb (OB) output neurons [mitral/tufted cells (MCs)] are temporally transformed relative to those of sensory neurons. We investigated how OB circuits shape inhalation-driven dynamics in MCs using a modeling approach that was highly constrained by experimental results. First, we constructed models of canonical OB circuits that included mono- and disynaptic feedforward excitation, recurrent inhibition and feedforward inhibition of the MC. We then used experimental data to drive inputs to the models and to tune parameters; inputs were derived from sensory neuron responses during natural odorant sampling (sniffing) in awake rats, and model output was compared with recordings of MC responses to odorants sampled with the same sniff waveforms. This approach allowed us to identify OB circuit features underlying the temporal transformation of sensory inputs into inhalation-linked patterns of MC spike output. We found that realistic input-output transformations can be achieved independently by multiple circuits, including feedforward inhibition with slow onset and decay kinetics and parallel feedforward MC excitation mediated by external tufted cells. We also found that recurrent and feedforward inhibition had differential impacts on MC firing rates and on inhalation-linked response dynamics. These results highlight the importance of investigating neural circuits in a naturalistic context and provide a framework for further explorations of signal processing by OB networks. PMID:25717156

Robust design of feedback feed-forward iterative learning control based on 2D system theory for linear uncertain systems

NASA Astrophysics Data System (ADS)

Li, Zhifu; Hu, Yueming; Li, Di

2016-08-01

For a class of linear discrete-time uncertain systems, a feedback feed-forward iterative learning control (ILC) scheme is proposed, which is comprised of an iterative learning controller and two current iteration feedback controllers. The iterative learning controller is used to improve the performance along the iteration direction and the feedback controllers are used to improve the performance along the time direction. First of all, the uncertain feedback feed-forward ILC system is presented by an uncertain two-dimensional Roesser model system. Then, two robust control schemes are proposed. One can ensure that the feedback feed-forward ILC system is bounded-input bounded-output stable along time direction, and the other can ensure that the feedback feed-forward ILC system is asymptotically stable along time direction. Both schemes can guarantee the system is robust monotonically convergent along the iteration direction. Third, the robust convergent sufficient conditions are given, which contains a linear matrix inequality (LMI). Moreover, the LMI can be used to determine the gain matrix of the feedback feed-forward iterative learning controller. Finally, the simulation results are presented to demonstrate the effectiveness of the proposed schemes.

Feedforward-feedback control of dissolved oxygen concentration in a predenitrification system.

PubMed

Yong, Ma; Yongzhen, Peng; Shuying, Wang

2005-07-01

As the largest single energy-consuming component in most biological wastewater treatment systems, aeration control is of great interest from the point of view of saving energy and improving wastewater treatment plant efficiency. In this paper, three different strategies, including conventional constant dissolved oxygen (DO) set-point control, cascade DO set-point control, and feedforward-feedback DO set-point control were evaluated using the denitrification layout of the IWA simulation benchmark. Simulation studies showed that the feedforward-feedback DO set-point control strategy was better than the other control strategies at meeting the effluent standards and reducing operational costs. The control strategy works primarily by feedforward control based on an ammonium sensor located at the head of the aerobic process. It has an important advantage over effluent measurements in that there is no (or only a very short) time delay for information; feedforward control was combined with slow feedback control to compensate for model approximations. The feedforward-feedback DO control was implemented in a lab-scale wastewater treatment plant for a period of 60 days. Compared to operation with constant DO concentration, the required airflow could be reduced by up to 8-15% by employing the feedforward-feedback DO-control strategy, and the effluent ammonia concentration could be reduced by up to 15-25%. This control strategy can be expected to be accepted by the operating personnel in wastewater treatment plants.

Valid approximation of spatially distributed grain size distributions - A priori information encoded to a feedforward network

NASA Astrophysics Data System (ADS)

Berthold, T.; Milbradt, P.; Berkhahn, V.

2018-04-01

This paper presents a model for the approximation of multiple, spatially distributed grain size distributions based on a feedforward neural network. Since a classical feedforward network does not guarantee to produce valid cumulative distribution functions, a priori information is incor porated into the model by applying weight and architecture constraints. The model is derived in two steps. First, a model is presented that is able to produce a valid distribution function for a single sediment sample. Although initially developed for sediment samples, the model is not limited in its application; it can also be used to approximate any other multimodal continuous distribution function. In the second part, the network is extended in order to capture the spatial variation of the sediment samples that have been obtained from 48 locations in the investigation area. Results show that the model provides an adequate approximation of grain size distributions, satisfying the requirements of a cumulative distribution function.

Prediction Analysis of Weld-Bead and Heat Affected Zone in TIG welding using Artificial Neural Networks

NASA Astrophysics Data System (ADS)

Saldanha, Shamith L.; Kalaichelvi, V.; Karthikeyan, R.

2018-04-01

TIG Welding is a high quality form of welding which is very popular in industries. It is one of the few types of welding that can be used to join dissimilar metals. Here a weld joint is formed between stainless steel and monel alloy. It is desired to have control over the weld geometry of such a joint through the adjustment of experimental parameters which are welding current, wire feed speed, arc length and the shielding gas flow rate. To facilitate the automation of the same, a model of the welding system is needed. However the underlying welding process is complex and non-linear, and analytical methods are impractical for industrial use. Therefore artificial neural networks (ANN) are explored for developing the model, as they are well-suited for modelling non-linear multi-variate data. Feed-forward neural networks with backpropagation training algorithm are used, and the data for training the ANN taken from experimental work. There are four outputs corresponding to the weld geometry. Different training and testing phases were carried out using MATLAB software and ANN approximates the given data with minimum amount of error.

Prediction of hot deformation behavior of high phosphorus steel using artificial neural network

NASA Astrophysics Data System (ADS)

Singh, Kanchan; Rajput, S. K.; Soota, T.; Verma, Vijay; Singh, Dharmendra

2018-03-01

To predict the hot deformation behavior of high phosphorus steel, the hot compression experiments were performed with the help of thermo-mechanical simulator Gleeble® 3800 in the temperatures ranging from 750 °C to 1050 °C and strain rates of 0.001 s-1, 0.01 s-1, 0.1 s-1, 0.5 s-1, 1.0 s-1 and 10 s-1. The experimental stress-strain data are employed to develop artificial neural network (ANN) model and their predictability. Using different combination of temperature, strain and strain rate as a input parameter and obtained experimental stress as a target, a multi-layer ANN model based on feed-forward back-propagation algorithm is trained, to predict the flow stress for a given processing condition. The relative error between predicted and experimental stress are in the range of ±3.5%, whereas the correlation coefficient (R2) of training and testing data are 0.99986 and 0.99999 respectively. This shows that a well-trained ANN model has excellent capability to predict the hot deformation behavior of materials. Comparative study shows quite good agreement of predicted and experimental values.

Video Feedforward for Reading

ERIC Educational Resources Information Center

Dowrick, Peter W.; Kim-Rupnow, Weol Soon; Power, Thomas J.

2006-01-01

Video feedforward can create images of positive futures, as has been shown by researchers using self-modeling methods to teach new skills with carefully planned and edited videos that show the future capability of the individual. As a supplement to tutoring provided by community members, we extended these practices to young children struggling to…

An approach to multivariable control of manipulators

NASA Technical Reports Server (NTRS)

Seraji, H.

1987-01-01

The paper presents simple schemes for multivariable control of multiple-joint robot manipulators in joint and Cartesian coordinates. The joint control scheme consists of two independent multivariable feedforward and feedback controllers. The feedforward controller is the minimal inverse of the linearized model of robot dynamics and contains only proportional-double-derivative (PD2) terms - implying feedforward from the desired position, velocity and acceleration. This controller ensures that the manipulator joint angles track any reference trajectories. The feedback controller is of proportional-integral-derivative (PID) type and is designed to achieve pole placement. This controller reduces any initial tracking error to zero as desired and also ensures that robust steady-state tracking of step-plus-exponential trajectories is achieved by the joint angles. Simple and explicit expressions of computation of the feedforward and feedback gains are obtained based on the linearized model of robot dynamics. This leads to computationally efficient schemes for either on-line gain computation or off-line gain scheduling to account for variations in the linearized robot model due to changes in the operating point. The joint control scheme is extended to direct control of the end-effector motion in Cartesian space. Simulation results are given for illustration.

A multifactorial analysis of obesity as CVD risk factor: use of neural network based methods in a nutrigenetics context.

PubMed

Valavanis, Ioannis K; Mougiakakou, Stavroula G; Grimaldi, Keith A; Nikita, Konstantina S

2010-09-08

Obesity is a multifactorial trait, which comprises an independent risk factor for cardiovascular disease (CVD). The aim of the current work is to study the complex etiology beneath obesity and identify genetic variations and/or factors related to nutrition that contribute to its variability. To this end, a set of more than 2300 white subjects who participated in a nutrigenetics study was used. For each subject a total of 63 factors describing genetic variants related to CVD (24 in total), gender, and nutrition (38 in total), e.g. average daily intake in calories and cholesterol, were measured. Each subject was categorized according to body mass index (BMI) as normal (BMI ≤ 25) or overweight (BMI > 25). Two artificial neural network (ANN) based methods were designed and used towards the analysis of the available data. These corresponded to i) a multi-layer feed-forward ANN combined with a parameter decreasing method (PDM-ANN), and ii) a multi-layer feed-forward ANN trained by a hybrid method (GA-ANN) which combines genetic algorithms and the popular back-propagation training algorithm. PDM-ANN and GA-ANN were comparatively assessed in terms of their ability to identify the most important factors among the initial 63 variables describing genetic variations, nutrition and gender, able to classify a subject into one of the BMI related classes: normal and overweight. The methods were designed and evaluated using appropriate training and testing sets provided by 3-fold Cross Validation (3-CV) resampling. Classification accuracy, sensitivity, specificity and area under receiver operating characteristics curve were utilized to evaluate the resulted predictive ANN models. The most parsimonious set of factors was obtained by the GA-ANN method and included gender, six genetic variations and 18 nutrition-related variables. The corresponding predictive model was characterized by a mean accuracy equal of 61.46% in the 3-CV testing sets. The ANN based methods revealed factors that interactively contribute to obesity trait and provided predictive models with a promising generalization ability. In general, results showed that ANNs and their hybrids can provide useful tools for the study of complex traits in the context of nutrigenetics.

Using Upper Layer Weights to Efficiently Construct and Train Feedforward Neural Networks Executing Backpropagation

DTIC Science & Technology

2011-03-01

algorithm is utilized by Belue, Steppe, & Bauer and Kocur , et al. (Belue, Steppe, & Bauer, April 1996) ( Kocur , et al., 1996). Bacauskiene and...Society. Cardiff, UK. Kocur , C., Roger, S., Myers, L., Burns, T., Hoffmeister, J., Bauer, K., et al. (1996). Using neural networks to select

Tactile feedback is an effective instrument for the training of grasping with a prosthesis at low- and medium-force levels.

PubMed

De Nunzio, Alessandro Marco; Dosen, Strahinja; Lemling, Sabrina; Markovic, Marko; Schweisfurth, Meike Annika; Ge, Nan; Graimann, Bernhard; Falla, Deborah; Farina, Dario

2017-08-01

Grasping is a complex task routinely performed in an anticipatory (feedforward) manner, where sensory feedback is responsible for learning and updating the internal model of grasp dynamics. This study aims at evaluating whether providing a proportional tactile force feedback during the myoelectric control of a prosthesis facilitates learning a stable internal model of the prosthesis force control. Ten able-bodied subjects controlled a sensorized myoelectric prosthesis performing four blocks of consecutive grasps at three levels of target force (30, 50, and 70%), repeatedly closing the fully opened hand. In the first and third block, the subjects received tactile and visual feedback, respectively, while during the second and fourth block, the feedback was removed. The subjects also performed an additional block with no feedback 1 day after the training (Retest). The median and interquartile range of the generated forces was computed to assess the accuracy and precision of force control. The results demonstrated that the feedback was indeed an effective instrument for the training of prosthesis control. After the training, the subjects were still able to accurately generate the desired force for the low and medium target (30 and 50% of maximum force available in a prosthesis), despite the feedback being removed within the session and during the retest (low target force). However, the training was substantially less successful for high forces (70% of prosthesis maximum force), where subjects exhibited a substantial loss of accuracy as soon as the feedback was removed. The precision of control decreased with higher forces and it was consistent across conditions, determined by an intrinsic variability of repeated myoelectric grasping. This study demonstrated that the subject could rely on the tactile feedback to adjust the motor command to the prosthesis across trials. The subjects adjusted the mean level of muscle activation (accuracy), whereas the precision could not be modulated as it depends on the intrinsic myoelectric variability. They were also able to maintain the feedforward command even after the feedback was removed, demonstrating thereby a stable learning, but the retention depended on the level of the target force. This is an important insight into the role of feedback as an instrument for learning of anticipatory prosthesis force control.

Neural network computer simulation of medical aerosols.

PubMed

Richardson, C J; Barlow, D J

1996-06-01

Preliminary investigations have been conducted to assess the potential for using artificial neural networks to simulate aerosol behaviour, with a view to employing this type of methodology in the evaluation and design of pulmonary drug-delivery systems. Details are presented of the general purpose software developed for these tasks; it implements a feed-forward back-propagation algorithm with weight decay and connection pruning, the user having complete run-time control of the network architecture and mode of training. A series of exploratory investigations is then reported in which different network structures and training strategies are assessed in terms of their ability to simulate known patterns of fluid flow in simple model systems. The first of these involves simulations of cellular automata-generated data for fluid flow through a partially obstructed two-dimensional pipe. The artificial neural networks are shown to be highly successful in simulating the behaviour of this simple linear system, but with important provisos relating to the information content of the training data and the criteria used to judge when the network is properly trained. A second set of investigations is then reported in which similar networks are used to simulate patterns of fluid flow through aerosol generation devices, using training data furnished through rigorous computational fluid dynamics modelling. These more complex three-dimensional systems are modelled with equal success. It is concluded that carefully tailored, well trained networks could provide valuable tools not just for predicting but also for analysing the spatial dynamics of pharmaceutical aerosols.

Hybrid Analytical and Data-Driven Modeling for Feed-Forward Robot Control †

PubMed Central

Reinhart, René Felix; Shareef, Zeeshan; Steil, Jochen Jakob

2017-01-01

Feed-forward model-based control relies on models of the controlled plant, e.g., in robotics on accurate knowledge of manipulator kinematics or dynamics. However, mechanical and analytical models do not capture all aspects of a plant’s intrinsic properties and there remain unmodeled dynamics due to varying parameters, unmodeled friction or soft materials. In this context, machine learning is an alternative suitable technique to extract non-linear plant models from data. However, fully data-based models suffer from inaccuracies as well and are inefficient if they include learning of well known analytical models. This paper thus argues that feed-forward control based on hybrid models comprising an analytical model and a learned error model can significantly improve modeling accuracy. Hybrid modeling here serves the purpose to combine the best of the two modeling worlds. The hybrid modeling methodology is described and the approach is demonstrated for two typical problems in robotics, i.e., inverse kinematics control and computed torque control. The former is performed for a redundant soft robot and the latter for a rigid industrial robot with redundant degrees of freedom, where a complete analytical model is not available for any of the platforms. PMID:28208697

Hybrid Analytical and Data-Driven Modeling for Feed-Forward Robot Control.

PubMed

Reinhart, René Felix; Shareef, Zeeshan; Steil, Jochen Jakob

2017-02-08

Feed-forward model-based control relies on models of the controlled plant, e.g., in robotics on accurate knowledge of manipulator kinematics or dynamics. However, mechanical and analytical models do not capture all aspects of a plant's intrinsic properties and there remain unmodeled dynamics due to varying parameters, unmodeled friction or soft materials. In this context, machine learning is an alternative suitable technique to extract non-linear plant models from data. However, fully data-based models suffer from inaccuracies as well and are inefficient if they include learning of well known analytical models. This paper thus argues that feed-forward control based on hybrid models comprising an analytical model and a learned error model can significantly improve modeling accuracy. Hybrid modeling here serves the purpose to combine the best of the two modeling worlds. The hybrid modeling methodology is described and the approach is demonstrated for two typical problems in robotics, i.e., inverse kinematics control and computed torque control. The former is performed for a redundant soft robot and the latter for a rigid industrial robot with redundant degrees of freedom, where a complete analytical model is not available for any of the platforms.

Feedforward compensation control of rotor imbalance for high-speed magnetically suspended centrifugal compressors using a novel adaptive notch filter

NASA Astrophysics Data System (ADS)

Zheng, Shiqiang; Feng, Rui

2016-03-01

This paper introduces a feedforward control strategy combined with a novel adaptive notch filter to solve the problem of rotor imbalance in high-speed Magnetically Suspended Centrifugal Compressors (MSCCs). Unbalance vibration force of rotor in MSCC is mainly composed of current stiffness force and displacement stiffness force. In this paper, the mathematical model of the unbalance vibration with the proportional-integral-derivative (PID) control laws is presented. In order to reduce the unbalance vibration, a novel adaptive notch filter is proposed to identify the synchronous frequency displacement of the rotor as a compensation signal to eliminate the current stiffness force. In addition, a feedforward channel from position component to control output is introduced to compensate displacement stiffness force to achieve a better performance. A simplified inverse model of power amplifier is included in the feedforward channel to reject the degrade performance caused by its low-pass characteristic. Simulation and experimental results on a MSCC demonstrate a significant effect on the synchronous vibration suppression of the magnetically suspended rotor at a high speed.

Using soft computing techniques to predict corrected air permeability using Thomeer parameters, air porosity and grain density

NASA Astrophysics Data System (ADS)

Nooruddin, Hasan A.; Anifowose, Fatai; Abdulraheem, Abdulazeez

2014-03-01

Soft computing techniques are recently becoming very popular in the oil industry. A number of computational intelligence-based predictive methods have been widely applied in the industry with high prediction capabilities. Some of the popular methods include feed-forward neural networks, radial basis function network, generalized regression neural network, functional networks, support vector regression and adaptive network fuzzy inference system. A comparative study among most popular soft computing techniques is presented using a large dataset published in literature describing multimodal pore systems in the Arab D formation. The inputs to the models are air porosity, grain density, and Thomeer parameters obtained using mercury injection capillary pressure profiles. Corrected air permeability is the target variable. Applying developed permeability models in recent reservoir characterization workflow ensures consistency between micro and macro scale information represented mainly by Thomeer parameters and absolute permeability. The dataset was divided into two parts with 80% of data used for training and 20% for testing. The target permeability variable was transformed to the logarithmic scale as a pre-processing step and to show better correlations with the input variables. Statistical and graphical analysis of the results including permeability cross-plots and detailed error measures were created. In general, the comparative study showed very close results among the developed models. The feed-forward neural network permeability model showed the lowest average relative error, average absolute relative error, standard deviations of error and root means squares making it the best model for such problems. Adaptive network fuzzy inference system also showed very good results.

Linear quadratic Gaussian and feedforward controllers for the DSS-13 antenna

NASA Technical Reports Server (NTRS)

Gawronski, W. K.; Racho, C. S.; Mellstrom, J. A.

1994-01-01

The controller development and the tracking performance evaluation for the DSS-13 antenna are presented. A trajectory preprocessor, linear quadratic Gaussian (LQG) controller, feedforward controller, and their combination were designed, built, analyzed, and tested. The antenna exhibits nonlinear behavior when the input to the antenna and/or the derivative of this input exceeds the imposed limits; for slewing and acquisition commands, these limits are typically violated. A trajectory preprocessor was designed to ensure that the antenna behaves linearly, just to prevent nonlinear limit cycling. The estimator model for the LQG controller was identified from the data obtained from the field test. Based on an LQG balanced representation, a reduced-order LQG controller was obtained. The feedforward controller and the combination of the LQG and feedforward controller were also investigated. The performance of the controllers was evaluated with the tracking errors (due to following a trajectory) and the disturbance errors (due to the disturbances acting on the antenna). The LQG controller has good disturbance rejection properties and satisfactory tracking errors. The feedforward controller has small tracking errors but poor disturbance rejection properties. The combined LQG and feedforward controller exhibits small tracking errors as well as good disturbance rejection properties. However, the cost for this performance is the complexity of the controller.

Constitutive flow behaviour of austenitic stainless steels under hot deformation: artificial neural network modelling to understand, evaluate and predict

NASA Astrophysics Data System (ADS)

Mandal, Sumantra; Sivaprasad, P. V.; Venugopal, S.; Murthy, K. P. N.

2006-09-01

An artificial neural network (ANN) model is developed to predict the constitutive flow behaviour of austenitic stainless steels during hot deformation. The input parameters are alloy composition and process variables whereas flow stress is the output. The model is based on a three-layer feed-forward ANN with a back-propagation learning algorithm. The neural network is trained with an in-house database obtained from hot compression tests on various grades of austenitic stainless steels. The performance of the model is evaluated using a wide variety of statistical indices. Good agreement between experimental and predicted data is obtained. The correlation between individual alloying elements and high temperature flow behaviour is investigated by employing the ANN model. The results are found to be consistent with the physical phenomena. The model can be used as a guideline for new alloy development.

Synaptic convergence regulates synchronization-dependent spike transfer in feedforward neural networks.

PubMed

Sailamul, Pachaya; Jang, Jaeson; Paik, Se-Bum

2017-12-01

Correlated neural activities such as synchronizations can significantly alter the characteristics of spike transfer between neural layers. However, it is not clear how this synchronization-dependent spike transfer can be affected by the structure of convergent feedforward wiring. To address this question, we implemented computer simulations of model neural networks: a source and a target layer connected with different types of convergent wiring rules. In the Gaussian-Gaussian (GG) model, both the connection probability and the strength are given as Gaussian distribution as a function of spatial distance. In the Uniform-Constant (UC) and Uniform-Exponential (UE) models, the connection probability density is a uniform constant within a certain range, but the connection strength is set as a constant value or an exponentially decaying function, respectively. Then we examined how the spike transfer function is modulated under these conditions, while static or synchronized input patterns were introduced to simulate different levels of feedforward spike synchronization. We observed that the synchronization-dependent modulation of the transfer function appeared noticeably different for each convergence condition. The modulation of the spike transfer function was largest in the UC model, and smallest in the UE model. Our analysis showed that this difference was induced by the different spike weight distributions that was generated from convergent synapses in each model. Our results suggest that, the structure of the feedforward convergence is a crucial factor for correlation-dependent spike control, thus must be considered important to understand the mechanism of information transfer in the brain.

On the use of harmony search algorithm in the training of wavelet neural networks

NASA Astrophysics Data System (ADS)

Lai, Kee Huong; Zainuddin, Zarita; Ong, Pauline

2015-10-01

Wavelet neural networks (WNNs) are a class of feedforward neural networks that have been used in a wide range of industrial and engineering applications to model the complex relationships between the given inputs and outputs. The training of WNNs involves the configuration of the weight values between neurons. The backpropagation training algorithm, which is a gradient-descent method, can be used for this training purpose. Nonetheless, the solutions found by this algorithm often get trapped at local minima. In this paper, a harmony search-based algorithm is proposed for the training of WNNs. The training of WNNs, thus can be formulated as a continuous optimization problem, where the objective is to maximize the overall classification accuracy. Each candidate solution proposed by the harmony search algorithm represents a specific WNN architecture. In order to speed up the training process, the solution space is divided into disjoint partitions during the random initialization step of harmony search algorithm. The proposed training algorithm is tested onthree benchmark problems from the UCI machine learning repository, as well as one real life application, namely, the classification of electroencephalography signals in the task of epileptic seizure detection. The results obtained show that the proposed algorithm outperforms the traditional harmony search algorithm in terms of overall classification accuracy.

A continually online-trained neural network controller for brushless DC motor drives

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

Rubaai, A.; Kotaru, R.; Kankam, M.D.

2000-04-01

In this paper, a high-performance controller with simultaneous online identification and control is designed for brushless dc motor drives. The dynamics of the motor/load are modeled online, and controlled using two different neural network based identification and control schemes, as the system is in operation. In the first scheme, an attempt is made to control the rotor angular speed, utilizing a single three-hidden-layer network. The second scheme attempts to control the stator currents, using a predetermined control law as a function of the estimated states. This schemes incorporates three multilayered feedforward neural networks that are online trained, using the Levenburg-Marquadtmore » training algorithm. The control of the direct and quadrature components of the stator current successfully tracked a wide variety of trajectories after relatively short online training periods. The control strategy adapts to the uncertainties of the motor/load dynamics and, in addition, learns their inherent nonlinearities. Simulation results illustrated that a neurocontroller used in conjunction with adaptive control schemes can result in a flexible control device which may be utilized in a wide range of environments.« less

A P2P Botnet detection scheme based on decision tree and adaptive multilayer neural networks.

PubMed

Alauthaman, Mohammad; Aslam, Nauman; Zhang, Li; Alasem, Rafe; Hossain, M A

2018-01-01

In recent years, Botnets have been adopted as a popular method to carry and spread many malicious codes on the Internet. These malicious codes pave the way to execute many fraudulent activities including spam mail, distributed denial-of-service attacks and click fraud. While many Botnets are set up using centralized communication architecture, the peer-to-peer (P2P) Botnets can adopt a decentralized architecture using an overlay network for exchanging command and control data making their detection even more difficult. This work presents a method of P2P Bot detection based on an adaptive multilayer feed-forward neural network in cooperation with decision trees. A classification and regression tree is applied as a feature selection technique to select relevant features. With these features, a multilayer feed-forward neural network training model is created using a resilient back-propagation learning algorithm. A comparison of feature set selection based on the decision tree, principal component analysis and the ReliefF algorithm indicated that the neural network model with features selection based on decision tree has a better identification accuracy along with lower rates of false positives. The usefulness of the proposed approach is demonstrated by conducting experiments on real network traffic datasets. In these experiments, an average detection rate of 99.08 % with false positive rate of 0.75 % was observed.

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.

Where Do Features Come From?

ERIC Educational Resources Information Center

Hinton, Geoffrey

2014-01-01

It is possible to learn multiple layers of non-linear features by backpropagating error derivatives through a feedforward neural network. This is a very effective learning procedure when there is a huge amount of labeled training data, but for many learning tasks very few labeled examples are available. In an effort to overcome the need for…

Reformulated Neural Network (ReNN): a New Alternative for Data-driven Modelling in Hydrology and Water Resources Engineering

NASA Astrophysics Data System (ADS)

Razavi, S.; Tolson, B.; Burn, D.; Seglenieks, F.

2012-04-01

Reformulated Neural Network (ReNN) has been recently developed as an efficient and more effective alternative to feedforward multi-layer perceptron (MLP) neural networks [Razavi, S., and Tolson, B. A. (2011). "A new formulation for feedforward neural networks." IEEE Transactions on Neural Networks, 22(10), 1588-1598, DOI: 1510.1109/TNN.2011.2163169]. This presentation initially aims to introduce the ReNN to the water resources community and then demonstrates ReNN applications to water resources related problems. ReNN is essentially equivalent to a single-hidden-layer MLP neural network but defined on a new set of network variables which is more effective than the traditional set of network weights and biases. The main features of the new network variables are that they are geometrically interpretable and each variable has a distinct role in forming the network response. ReNN is more efficiently trained as it has a less complex error response surface. In addition to the ReNN training efficiency, the interpretability of the ReNN variables enables the users to monitor and understand the internal behaviour of the network while training. Regularization in the ReNN response can be also directly measured and controlled. This feature improves the generalization ability of the network. The appeal of the ReNN is demonstrated with two ReNN applications to water resources engineering problems. In the first application, the ReNN is used to model the rainfall-runoff relationships in multiple watersheds in the Great Lakes basin located in northeastern North America. Modelling inflows to the Great Lakes are of great importance to the management of the Great Lakes system. Due to the lack of some detailed physical data about existing control structures in many subwatersheds of this huge basin, the data-driven approach to modelling such as the ReNN are required to replace predictions from a physically-based rainfall runoff model. Unlike traditional MLPs, the ReNN does not necessarily require an independent set of data for validation as the ReNN has the capability to control and verify the network degree of regularization. As such, the ReNN can be very beneficial in this case study as the data available in this case study is limited. In the second application, ReNN is fitted on the response function of the SWAT hydrologic model to act as a fast-to-run response surface surrogate (i.e., metamodel) of the original computationally intensive SWAT model. Besides the training efficiency gains, the ReNN applications demonstrate how the ReNN interpretability could help users develop more reliable networks which perform predictably better in terms of generalization.

Feedforward and Feedback Control in Apraxia of Speech: Effects of Noise Masking on Vowel Production

ERIC Educational Resources Information Center

Maas, Edwin; Mailend, Marja-Liisa; Guenther, Frank H.

2015-01-01

Purpose: This study was designed to test two hypotheses about apraxia of speech (AOS) derived from the Directions Into Velocities of Articulators (DIVA) model (Guenther et al., 2006): the feedforward system deficit hypothesis and the feedback system deficit hypothesis. Method: The authors used noise masking to minimize auditory feedback during…

Video Feedforward for Rapid Learning of a Picture-Based Communication System

ERIC Educational Resources Information Center

Smith, Jemma; Hand, Linda; Dowrick, Peter W.

2014-01-01

This study examined the efficacy of video self modeling (VSM) using feedforward, to teach various goals of a picture exchange communication system (PECS). The participants were two boys with autism and one man with Down syndrome. All three participants were non-verbal with no current functional system of communication; the two children had long…

A training rule which guarantees finite-region stability for a class of closed-loop neural-network control systems.

PubMed

Kuntanapreeda, S; Fullmer, R R

1996-01-01

A training method for a class of neural network controllers is presented which guarantees closed-loop system stability. The controllers are assumed to be nonlinear, feedforward, sampled-data, full-state regulators implemented as single hidden-layer neural networks. The controlled systems must be locally hermitian and observable. Stability of the closed-loop system is demonstrated by determining a Lyapunov function, which can be used to identify a finite stability region about the regulator point.

Self-organizing linear output map (SOLO): An artificial neural network suitable for hydrologic modeling and analysis

NASA Astrophysics Data System (ADS)

Hsu, Kuo-Lin; Gupta, Hoshin V.; Gao, Xiaogang; Sorooshian, Soroosh; Imam, Bisher

2002-12-01

Artificial neural networks (ANNs) can be useful in the prediction of hydrologic variables, such as streamflow, particularly when the underlying processes have complex nonlinear interrelationships. However, conventional ANN structures suffer from network training issues that significantly limit their widespread application. This paper presents a multivariate ANN procedure entitled self-organizing linear output map (SOLO), whose structure has been designed for rapid, precise, and inexpensive estimation of network structure/parameters and system outputs. More important, SOLO provides features that facilitate insight into the underlying processes, thereby extending its usefulness beyond forecast applications as a tool for scientific investigations. These characteristics are demonstrated using a classic rainfall-runoff forecasting problem. Various aspects of model performance are evaluated in comparison with other commonly used modeling approaches, including multilayer feedforward ANNs, linear time series modeling, and conceptual rainfall-runoff modeling.

Adaptive online inverse control of a shape memory alloy wire actuator using a dynamic neural network

NASA Astrophysics Data System (ADS)

Mai, Huanhuan; Song, Gangbing; Liao, Xiaofeng

2013-01-01

Shape memory alloy (SMA) actuators exhibit severe hysteresis, a nonlinear behavior, which complicates control strategies and limits their applications. This paper presents a new approach to controlling an SMA actuator through an adaptive inverse model based controller that consists of a dynamic neural network (DNN) identifier, a copy dynamic neural network (CDNN) feedforward term and a proportional (P) feedback action. Unlike fixed hysteresis models used in most inverse controllers, the proposed one uses a DNN to identify online the relationship between the applied voltage to the actuator and the displacement (the inverse model). Even without a priori knowledge of the SMA hysteresis and without pre-training, the proposed controller can precisely control the SMA wire actuator in various tracking tasks by identifying online the inverse model of the SMA actuator. Experiments were conducted, and experimental results demonstrated real-time modeling capabilities of DNN and the performance of the adaptive inverse controller.

Simultaneous feedforward recruitment of the vasti in untrained postural tasks can be restored by physical therapy.

PubMed

Cowan, Sallie M; Bennell, Kim L; Hodges, Paul W; Crossley, Kay M; McConnell, Jenny

2003-05-01

Physical therapy rehabilitation strategies are commonly directed at the alteration of muscle recruitment in functional movements. The aim of this study was to investigate whether feedforward strategies of the vasti in people with patellofemoral pain syndrome can be changed by a physical therapy treatment program in a randomised, double blind, placebo controlled trial. Forty (25 female, 15 male) subjects aged 40 yrs or less (27.2+/-7.8 yrs). Subjects were allocated to either a placebo treatment or a physical therapy intervention program. The postural challenge used as the outcome measure was not included in the training program. Electromyography (EMG) onsets of vastus medialis obliquus (VMO), vastus lateralis (VL), tibialis anterior and soleus were assessed before and after the six week standardised treatment programs. At baseline the EMG onset of VL occurred prior to that of VMO in both subject groups. Following physical therapy intervention there was a significant change in the time of onset of EMG of VMO compared to VL with the onsets occurring simultaneously. This change was associated with a reduction in symptoms. In contrast, following placebo intervention the EMG onset of VL still occurred prior to that of VMO. The results indicate that the feedforward strategy used by the central nervous system to control the patella can be restored. Importantly, the data suggest that this intervention produced a change that was transferred to a task that was not specifically included in the training program. Furthermore, the change in motor control was associated with clinical improvement in symptoms.

Control of CA3 output by feedforward inhibition despite developmental changes in the excitation-inhibition balance.

PubMed

Torborg, Christine L; Nakashiba, Toshiaki; Tonegawa, Susumu; McBain, Chris J

2010-11-17

In somatosensory cortex, the relative balance of excitation and inhibition determines how effectively feedforward inhibition enforces the temporal fidelity of action potentials. Within the CA3 region of the hippocampus, glutamatergic mossy fiber (MF) synapses onto CA3 pyramidal cells (PCs) provide strong monosynaptic excitation that exhibit prominent facilitation during repetitive activity. We demonstrate in the juvenile CA3 that MF-driven polysynaptic IPSCs facilitate to maintain a fixed EPSC-IPSC ratio during short-term plasticity. In contrast, in young adult mice this MF-driven polysynaptic inhibitory input can facilitate or depress in response to short trains of activity. Transgenic mice lacking the feedback inhibitory loop continue to exhibit both facilitating and depressing polysynaptic IPSCs, indicating that this robust inhibition is not caused by the secondary engagement of feedback inhibition. Surprisingly, eliminating MF-driven inhibition onto CA3 pyramidal cells by blockade of GABA(A) receptors did not lead to a loss of temporal precision of the first action potential observed after a stimulus but triggered in many cases a long excitatory plateau potential capable of triggering repetitive action potential firing. These observations indicate that, unlike other regions of the brain, the temporal precision of single MF-driven action potentials is dictated primarily by the kinetics of MF EPSPs, not feedforward inhibition. Instead, feedforward inhibition provides a robust regulation of CA3 PC excitability across development to prevent excessive depolarization by the monosynaptic EPSP and multiple action potential firings.

Hybrid Feedforward-Feedback Noise Control Using Virtual Sensors

NASA Technical Reports Server (NTRS)

Bean, Jacob; Fuller, Chris; Schiller, Noah

2016-01-01

Several approaches to active noise control using virtual sensors are evaluated for eventual use in an active headrest. Specifically, adaptive feedforward, feedback, and hybrid control structures are compared. Each controller incorporates the traditional filtered-x least mean squares algorithm. The feedback controller is arranged in an internal model configuration to draw comparisons with standard feedforward control theory results. Simulation and experimental results are presented that illustrate each controllers ability to minimize the pressure at both physical and virtual microphone locations. The remote microphone technique is used to obtain pressure estimates at the virtual locations. It is shown that a hybrid controller offers performance benefits over the traditional feedforward and feedback controllers. Stability issues associated with feedback and hybrid controllers are also addressed. Experimental results show that 15-20 dB reduction in broadband disturbances can be achieved by minimizing the measured pressure, whereas 10-15 dB reduction is obtained when minimizing the estimated pressure at a virtual location.

Combined feedforward and feedback control of a redundant, nonlinear, dynamic musculoskeletal system.

PubMed

Blana, Dimitra; Kirsch, Robert F; Chadwick, Edward K

2009-05-01

A functional electrical stimulation controller is presented that uses a combination of feedforward and feedback for arm control in high-level injury. The feedforward controller generates the muscle activations nominally required for desired movements, and the feedback controller corrects for errors caused by muscle fatigue and external disturbances. The feedforward controller is an artificial neural network (ANN) which approximates the inverse dynamics of the arm. The feedback loop includes a PID controller in series with a second ANN representing the nonlinear properties and biomechanical interactions of muscles and joints. The controller was designed and tested using a two-joint musculoskeletal model of the arm that includes four mono-articular and two bi-articular muscles. Its performance during goal-oriented movements of varying amplitudes and durations showed a tracking error of less than 4 degrees in ideal conditions, and less than 10 degrees even in the case of considerable fatigue and external disturbances.

The Roles of Feedback and Feedforward as Humans Learn to Control Unknown Dynamic Systems.

PubMed

Zhang, Xingye; Wang, Shaoqian; Hoagg, Jesse B; Seigler, T Michael

2018-02-01

We present results from an experiment in which human subjects interact with an unknown dynamic system 40 times during a two-week period. During each interaction, subjects are asked to perform a command-following (i.e., pursuit tracking) task. Each subject's performance at that task improves from the first trial to the last trial. For each trial, we use subsystem identification to estimate each subject's feedforward (or anticipatory) control, feedback (or reactive) control, and feedback time delay. Over the 40 trials, the magnitudes of the identified feedback controllers and the identified feedback time delays do not change significantly. In contrast, the identified feedforward controllers do change significantly. By the last trial, the average identified feedforward controller approximates the inverse of the dynamic system. This observation provides evidence that a fundamental component of human learning is updating the anticipatory control until it models the inverse dynamics.

Slewing maneuvers and vibration control of space structures by feedforward/feedback moment-gyro controls

NASA Technical Reports Server (NTRS)

Yang, Li-Farn; Mikulas, Martin M., Jr.; Park, K. C.; Su, Renjeng

1993-01-01

This paper presents a moment-gyro control approach to the maneuver and vibration suppression of a flexible truss arm undergoing a constant slewing motion. The overall slewing motion is triggered by a feedforward input, and a companion feedback controller is employed to augment the feedforward input and subsequently to control vibrations. The feedforward input for the given motion requirement is determined from the combined CMG (Control Momentum Gyro) devices and the desired rigid-body motion. The rigid-body dynamic model has enabled us to identify the attendant CMG momentum saturation constraints. The task for vibration control is carried out in two stages; first in the search of a suitable CMG placement along the beam span for various slewing maneuvers, and subsequently in the development of Liapunov-based control algorithms for CMG spin-stabilization. Both analytical and numerical results are presented to show the effectiveness of the present approach.

Neural network based glucose - insulin metabolism models for children with Type 1 diabetes.

PubMed

Mougiakakou, Stavroula G; Prountzou, Aikaterini; Iliopoulou, Dimitra; Nikita, Konstantina S; Vazeou, Andriani; Bartsocas, Christos S

2006-01-01

In this paper two models for the simulation of glucose-insulin metabolism of children with Type 1 diabetes are presented. The models are based on the combined use of Compartmental Models (CMs) and artificial Neural Networks (NNs). Data from children with Type 1 diabetes, stored in a database, have been used as input to the models. The data are taken from four children with Type 1 diabetes and contain information about glucose levels taken from continuous glucose monitoring system, insulin intake and food intake, along with corresponding time. The influences of taken insulin on plasma insulin concentration, as well as the effect of food intake on glucose input into the blood from the gut, are estimated from the CMs. The outputs of CMs, along with previous glucose measurements, are fed to a NN, which provides short-term prediction of glucose values. For comparative reasons two different NN architectures have been tested: a Feed-Forward NN (FFNN) trained with the back-propagation algorithm with adaptive learning rate and momentum, and a Recurrent NN (RNN), trained with the Real Time Recurrent Learning (RTRL) algorithm. The results indicate that the best prediction performance can be achieved by the use of RNN.

Adaptive regulation of sparseness by feedforward inhibition

PubMed Central

Assisi, Collins; Stopfer, Mark; Laurent, Gilles; Bazhenov, Maxim

2014-01-01

In the mushroom body of insects, odors are represented by very few spikes in a small number of neurons, a highly efficient strategy known as sparse coding. Physiological studies of these neurons have shown that sparseness is maintained across thousand-fold changes in odor concentration. Using a realistic computational model, we propose that sparseness in the olfactory system is regulated by adaptive feedforward inhibition. When odor concentration changes, feedforward inhibition modulates the duration of the temporal window over which the mushroom body neurons may integrate excitatory presynaptic input. This simple adaptive mechanism could maintain the sparseness of sensory representations across wide ranges of stimulus conditions. PMID:17660812

Simulative and experimental investigation on stator winding turn and unbalanced supply voltage fault diagnosis in induction motors using Artificial Neural Networks.

PubMed

Lashkari, Negin; Poshtan, Javad; Azgomi, Hamid Fekri

2015-11-01

The three-phase shift between line current and phase voltage of induction motors can be used as an efficient fault indicator to detect and locate inter-turn stator short-circuit (ITSC) fault. However, unbalanced supply voltage is one of the contributing factors that inevitably affect stator currents and therefore the three-phase shift. Thus, it is necessary to propose a method that is able to identify whether the unbalance of three currents is caused by ITSC or supply voltage fault. This paper presents a feedforward multilayer-perceptron Neural Network (NN) trained by back propagation, based on monitoring negative sequence voltage and the three-phase shift. The data which are required for training and test NN are generated using simulated model of stator. The experimental results are presented to verify the superior accuracy of the proposed method. Copyright © 2015. Published by Elsevier Ltd.

Emerging feed-forward inhibition allows the robust formation of direction selectivity in the developing ferret visual cortex

PubMed Central

Escobar, Gina M.; Maffei, Arianna; Miller, Paul

2014-01-01

The computation of direction selectivity requires that a cell respond to joint spatial and temporal characteristics of the stimulus that cannot be separated into independent components. Direction selectivity in ferret visual cortex is not present at the time of eye opening but instead develops in the days and weeks following eye opening in a process that requires visual experience with moving stimuli. Classic Hebbian or spike timing-dependent modification of excitatory feed-forward synaptic inputs is unable to produce direction-selective cells from unselective or weakly directionally biased initial conditions because inputs eventually grow so strong that they can independently drive cortical neurons, violating the joint spatial-temporal activation requirement. Furthermore, without some form of synaptic competition, cells cannot develop direction selectivity in response to training with bidirectional stimulation, as cells in ferret visual cortex do. We show that imposing a maximum lateral geniculate nucleus (LGN)-to-cortex synaptic weight allows neurons to develop direction-selective responses that maintain the requirement for joint spatial and temporal activation. We demonstrate that a novel form of inhibitory plasticity, postsynaptic activity-dependent long-term potentiation of inhibition (POSD-LTPi), which operates in the developing cortex at the time of eye opening, can provide synaptic competition and enables robust development of direction-selective receptive fields with unidirectional or bidirectional stimulation. We propose a general model of the development of spatiotemporal receptive fields that consists of two phases: an experience-independent establishment of initial biases, followed by an experience-dependent amplification or modification of these biases via correlation-based plasticity of excitatory inputs that compete against gradually increasing feed-forward inhibition. PMID:24598528

Application of Neural Networks for classification of Patau, Edwards, Down, Turner and Klinefelter Syndrome based on first trimester maternal serum screening data, ultrasonographic findings and patient demographics.

PubMed

Catic, Aida; Gurbeta, Lejla; Kurtovic-Kozaric, Amina; Mehmedbasic, Senad; Badnjevic, Almir

2018-02-13

The usage of Artificial Neural Networks (ANNs) for genome-enabled classifications and establishing genome-phenotype correlations have been investigated more extensively over the past few years. The reason for this is that ANNs are good approximates of complex functions, so classification can be performed without the need for explicitly defined input-output model. This engineering tool can be applied for optimization of existing methods for disease/syndrome classification. Cytogenetic and molecular analyses are the most frequent tests used in prenatal diagnostic for the early detection of Turner, Klinefelter, Patau, Edwards and Down syndrome. These procedures can be lengthy, repetitive; and often employ invasive techniques so a robust automated method for classifying and reporting prenatal diagnostics would greatly help the clinicians with their routine work. The database consisted of data collected from 2500 pregnant woman that came to the Institute of Gynecology, Infertility and Perinatology "Mehmedbasic" for routine antenatal care between January 2000 and December 2016. During first trimester all women were subject to screening test where values of maternal serum pregnancy-associated plasma protein A (PAPP-A) and free beta human chorionic gonadotropin (β-hCG) were measured. Also, fetal nuchal translucency thickness and the presence or absence of the nasal bone was observed using ultrasound. The architectures of linear feedforward and feedback neural networks were investigated for various training data distributions and number of neurons in hidden layer. Feedback neural network architecture out performed feedforward neural network architecture in predictive ability for all five aneuploidy prenatal syndrome classes. Feedforward neural network with 15 neurons in hidden layer achieved classification sensitivity of 92.00%. Classification sensitivity of feedback (Elman's) neural network was 99.00%. Average accuracy of feedforward neural network was 89.6% and for feedback was 98.8%. The results presented in this paper prove that an expert diagnostic system based on neural networks can be efficiently used for classification of five aneuploidy syndromes, covered with this study, based on first trimester maternal serum screening data, ultrasonographic findings and patient demographics. Developed Expert System proved to be simple, robust, and powerful in properly classifying prenatal aneuploidy syndromes.

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].

Combined contributions of feedforward and feedback inputs to bottom-up attention

PubMed Central

Khorsand, Peyman; Moore, Tirin; Soltani, Alireza

2015-01-01

In order to deal with a large amount of information carried by visual inputs entering the brain at any given point in time, the brain swiftly uses the same inputs to enhance processing in one part of visual field at the expense of the others. These processes, collectively called bottom-up attentional selection, are assumed to solely rely on feedforward processing of the external inputs, as it is implied by the nomenclature. Nevertheless, evidence from recent experimental and modeling studies points to the role of feedback in bottom-up attention. Here, we review behavioral and neural evidence that feedback inputs are important for the formation of signals that could guide attentional selection based on exogenous inputs. Moreover, we review results from a modeling study elucidating mechanisms underlying the emergence of these signals in successive layers of neural populations and how they depend on feedback from higher visual areas. We use these results to interpret and discuss more recent findings that can further unravel feedforward and feedback neural mechanisms underlying bottom-up attention. We argue that while it is descriptively useful to separate feedforward and feedback processes underlying bottom-up attention, these processes cannot be mechanistically separated into two successive stages as they occur at almost the same time and affect neural activity within the same brain areas using similar neural mechanisms. Therefore, understanding the interaction and integration of feedforward and feedback inputs is crucial for better understanding of bottom-up attention. PMID:25784883

Neural net diagnostics for VLSI test

NASA Technical Reports Server (NTRS)

Lin, T.; Tseng, H.; Wu, A.; Dogan, N.; Meador, J.

1990-01-01

This paper discusses the application of neural network pattern analysis algorithms to the IC fault diagnosis problem. A fault diagnostic is a decision rule combining what is known about an ideal circuit test response with information about how it is distorted by fabrication variations and measurement noise. The rule is used to detect fault existence in fabricated circuits using real test equipment. Traditional statistical techniques may be used to achieve this goal, but they can employ unrealistic a priori assumptions about measurement data. Our approach to this problem employs an adaptive pattern analysis technique based on feedforward neural networks. During training, a feedforward network automatically captures unknown sample distributions. This is important because distributions arising from the nonlinear effects of process variation can be more complex than is typically assumed. A feedforward network is also able to extract measurement features which contribute significantly to making a correct decision. Traditional feature extraction techniques employ matrix manipulations which can be particularly costly for large measurement vectors. In this paper we discuss a software system which we are developing that uses this approach. We also provide a simple example illustrating the use of the technique for fault detection in an operational amplifier.

Combined feedforward and model-assisted active disturbance rejection control for non-minimum phase system.

PubMed

Sun, Li; Li, Donghai; Gao, Zhiqiang; Yang, Zhao; Zhao, Shen

2016-09-01

Control of the non-minimum phase (NMP) system is challenging, especially in the presence of modelling uncertainties and external disturbances. To this end, this paper presents a combined feedforward and model-assisted Active Disturbance Rejection Control (MADRC) strategy. Based on the nominal model, the feedforward controller is used to produce a tracking performance that has minimum settling time subject to a prescribed undershoot constraint. On the other hand, the unknown disturbances and uncertain dynamics beyond the nominal model are compensated by MADRC. Since the conventional Extended State Observer (ESO) is not suitable for the NMP system, a model-assisted ESO (MESO) is proposed based on the nominal observable canonical form. The convergence of MESO is proved in time domain. The stability, steady-state characteristics and robustness of the closed-loop system are analyzed in frequency domain. The proposed strategy has only one tuning parameter, i.e., the bandwidth of MESO, which can be readily determined with a prescribed robustness level. Some comparative examples are given to show the efficacy of the proposed method. This paper depicts a promising prospect of the model-assisted ADRC in dealing with complex systems. Copyright © 2016 ISA. Published by Elsevier Ltd. All rights reserved.

Integrate-and-fire vs Poisson models of LGN input to V1 cortex: noisier inputs reduce orientation selectivity

PubMed Central

Lin, I-Chun; Xing, Dajun; Shapley, Robert

2014-01-01

One of the reasons the visual cortex has attracted the interest of computational neuroscience is that it has well-defined inputs. The lateral geniculate nucleus (LGN) of the thalamus is the source of visual signals to the primary visual cortex (V1). Most large-scale cortical network models approximate the spike trains of LGN neurons as simple Poisson point processes. However, many studies have shown that neurons in the early visual pathway are capable of spiking with high temporal precision and their discharges are not Poisson-like. To gain an understanding of how response variability in the LGN influences the behavior of V1, we study response properties of model V1 neurons that receive purely feedforward inputs from LGN cells modeled either as noisy leaky integrate-and-fire (NLIF) neurons or as inhomogeneous Poisson processes. We first demonstrate that the NLIF model is capable of reproducing many experimentally observed statistical properties of LGN neurons. Then we show that a V1 model in which the LGN input to a V1 neuron is modeled as a group of NLIF neurons produces higher orientation selectivity than the one with Poisson LGN input. The second result implies that statistical characteristics of LGN spike trains are important for V1's function. We conclude that physiologically motivated models of V1 need to include more realistic LGN spike trains that are less noisy than inhomogeneous Poisson processes. PMID:22684587

Integrate-and-fire vs Poisson models of LGN input to V1 cortex: noisier inputs reduce orientation selectivity.

PubMed

Lin, I-Chun; Xing, Dajun; Shapley, Robert

2012-12-01

One of the reasons the visual cortex has attracted the interest of computational neuroscience is that it has well-defined inputs. The lateral geniculate nucleus (LGN) of the thalamus is the source of visual signals to the primary visual cortex (V1). Most large-scale cortical network models approximate the spike trains of LGN neurons as simple Poisson point processes. However, many studies have shown that neurons in the early visual pathway are capable of spiking with high temporal precision and their discharges are not Poisson-like. To gain an understanding of how response variability in the LGN influences the behavior of V1, we study response properties of model V1 neurons that receive purely feedforward inputs from LGN cells modeled either as noisy leaky integrate-and-fire (NLIF) neurons or as inhomogeneous Poisson processes. We first demonstrate that the NLIF model is capable of reproducing many experimentally observed statistical properties of LGN neurons. Then we show that a V1 model in which the LGN input to a V1 neuron is modeled as a group of NLIF neurons produces higher orientation selectivity than the one with Poisson LGN input. The second result implies that statistical characteristics of LGN spike trains are important for V1's function. We conclude that physiologically motivated models of V1 need to include more realistic LGN spike trains that are less noisy than inhomogeneous Poisson processes.

Video feedforward for rapid learning of a picture-based communication system.

PubMed

Smith, Jemma; Hand, Linda; Dowrick, Peter W

2014-04-01

This study examined the efficacy of video self modeling (VSM) using feedforward, to teach various goals of a picture exchange communication system (PECS). The participants were two boys with autism and one man with Down syndrome. All three participants were non-verbal with no current functional system of communication; the two children had long histories of PECS failure. A series of replications, with different length baselines, was used to examine whether video self modeling could replace the PECS method of teaching to achieve the same goals. All three participants showed rapid learning of their target behavior when introduced to their self modeling videos, and effects generalized without the need for further intervention. We conclude that VSM, using feedforward, can provide a fast, simple way of teaching the use of a picture-based communication system without the need for prompts or intensive operant conditioning. VSM may provide an accessible, easy-to-use alternative to common methods of teaching augmentative and alternative communication systems.

Hebbian Learning in a Random Network Captures Selectivity Properties of the Prefrontal Cortex.

PubMed

Lindsay, Grace W; Rigotti, Mattia; Warden, Melissa R; Miller, Earl K; Fusi, Stefano

2017-11-08

Complex cognitive behaviors, such as context-switching and rule-following, are thought to be supported by the prefrontal cortex (PFC). Neural activity in the PFC must thus be specialized to specific tasks while retaining flexibility. Nonlinear "mixed" selectivity is an important neurophysiological trait for enabling complex and context-dependent behaviors. Here we investigate (1) the extent to which the PFC exhibits computationally relevant properties, such as mixed selectivity, and (2) how such properties could arise via circuit mechanisms. We show that PFC cells recorded from male and female rhesus macaques during a complex task show a moderate level of specialization and structure that is not replicated by a model wherein cells receive random feedforward inputs. While random connectivity can be effective at generating mixed selectivity, the data show significantly more mixed selectivity than predicted by a model with otherwise matched parameters. A simple Hebbian learning rule applied to the random connectivity, however, increases mixed selectivity and enables the model to match the data more accurately. To explain how learning achieves this, we provide analysis along with a clear geometric interpretation of the impact of learning on selectivity. After learning, the model also matches the data on measures of noise, response density, clustering, and the distribution of selectivities. Of two styles of Hebbian learning tested, the simpler and more biologically plausible option better matches the data. These modeling results provide clues about how neural properties important for cognition can arise in a circuit and make clear experimental predictions regarding how various measures of selectivity would evolve during animal training. SIGNIFICANCE STATEMENT The prefrontal cortex is a brain region believed to support the ability of animals to engage in complex behavior. How neurons in this area respond to stimuli-and in particular, to combinations of stimuli ("mixed selectivity")-is a topic of interest. Even though models with random feedforward connectivity are capable of creating computationally relevant mixed selectivity, such a model does not match the levels of mixed selectivity seen in the data analyzed in this study. Adding simple Hebbian learning to the model increases mixed selectivity to the correct level and makes the model match the data on several other relevant measures. This study thus offers predictions on how mixed selectivity and other properties evolve with training. Copyright © 2017 the authors 0270-6474/17/3711021-16$15.00/0.

Hebbian Learning in a Random Network Captures Selectivity Properties of the Prefrontal Cortex

PubMed Central

Lindsay, Grace W.

2017-01-01

Complex cognitive behaviors, such as context-switching and rule-following, are thought to be supported by the prefrontal cortex (PFC). Neural activity in the PFC must thus be specialized to specific tasks while retaining flexibility. Nonlinear “mixed” selectivity is an important neurophysiological trait for enabling complex and context-dependent behaviors. Here we investigate (1) the extent to which the PFC exhibits computationally relevant properties, such as mixed selectivity, and (2) how such properties could arise via circuit mechanisms. We show that PFC cells recorded from male and female rhesus macaques during a complex task show a moderate level of specialization and structure that is not replicated by a model wherein cells receive random feedforward inputs. While random connectivity can be effective at generating mixed selectivity, the data show significantly more mixed selectivity than predicted by a model with otherwise matched parameters. A simple Hebbian learning rule applied to the random connectivity, however, increases mixed selectivity and enables the model to match the data more accurately. To explain how learning achieves this, we provide analysis along with a clear geometric interpretation of the impact of learning on selectivity. After learning, the model also matches the data on measures of noise, response density, clustering, and the distribution of selectivities. Of two styles of Hebbian learning tested, the simpler and more biologically plausible option better matches the data. These modeling results provide clues about how neural properties important for cognition can arise in a circuit and make clear experimental predictions regarding how various measures of selectivity would evolve during animal training. SIGNIFICANCE STATEMENT The prefrontal cortex is a brain region believed to support the ability of animals to engage in complex behavior. How neurons in this area respond to stimuli—and in particular, to combinations of stimuli (“mixed selectivity”)—is a topic of interest. Even though models with random feedforward connectivity are capable of creating computationally relevant mixed selectivity, such a model does not match the levels of mixed selectivity seen in the data analyzed in this study. Adding simple Hebbian learning to the model increases mixed selectivity to the correct level and makes the model match the data on several other relevant measures. This study thus offers predictions on how mixed selectivity and other properties evolve with training. PMID:28986463

Adaptive positive position feedback control with a feedforward compensator of a magnetostrictive beam for vibration suppression

NASA Astrophysics Data System (ADS)

Bian, Leixiang; Zhu, Wei

2018-07-01

In this paper, a Fe–Ga alloy magnetostrictive beam is designed as an actuator to restrain the vibration of a supported mass. Dynamic modeling of the system based on the transfer matrix method of multibody system is first shown, and then a hybrid controller is developed to achieve vibration control. The proposed vibration controller combines a multi-mode adaptive positive position feedback (APPF) with a feedforward compensator. In the APPF control, an adaptive natural frequency estimator based on the recursive least-square method is developed to be used. In the feedforward compensator, the hysteresis of the magnetostrictive beam is linearized based on a Bouc–Wen model. The further remarkable vibration suppression capability of the proposed hybrid controller is demonstrated experimentally and compared with the positive position feedback controller. Experiment results show that the proposed controller is applicable to the magnetostrictive beam for improving vibration control effectiveness.

Feedback enhances feedforward figure-ground segmentation by changing firing mode.

PubMed

Supèr, Hans; Romeo, August

2011-01-01

In the visual cortex, feedback projections are conjectured to be crucial in figure-ground segregation. However, the precise function of feedback herein is unclear. Here we tested a hypothetical model of reentrant feedback. We used a previous developed 2-layered feedforward spiking network that is able to segregate figure from ground and included feedback connections. Our computer model data show that without feedback, neurons respond with regular low-frequency (∼9 Hz) bursting to a figure-ground stimulus. After including feedback the firing pattern changed into a regular (tonic) spiking pattern. In this state, we found an extra enhancement of figure responses and a further suppression of background responses resulting in a stronger figure-ground signal. Such push-pull effect was confirmed by comparing the figure-ground responses with the responses to a homogenous texture. We propose that feedback controls figure-ground segregation by influencing the neural firing patterns of feedforward projecting neurons.

Reconstruction of sub-surface archaeological remains from magnetic data using neural computing.

NASA Astrophysics Data System (ADS)

Bescoby, D. J.; Cawley, G. C.; Chroston, P. N.

2003-04-01

The remains of a former Roman colonial settlement, once part of the classical city of Butrint in southern Albania have been the subject of a high resolution magnetic survey using a caesium-vapour magnetometer. The survey revealed the surviving remains of an extensive planned settlement and a number of outlying buildings, today buried beneath over 0.5 m of alluvial deposits. The aim of the current research is to derive a sub-surface model from the magnetic survey measurements, allowing an enhanced archaeological interpretation of the data. Neural computing techniques are used to perform the non-linear mapping between magnetic data and corresponding sub-surface model parameters. The adoption of neural computing paradigms potentially holds several advantages over other modelling techniques, allowing fast solutions for complex data, while having a high tolerance to noise. A multi-layer perceptron network with a feed-forward architecture is trained to estimate the shape and burial depth of wall foundations using a series of representative models as training data. Parameters used to forward model the training data sets are derived from a number of trial trench excavations targeted over features identified by the magnetic survey. The training of the network was optimized by first applying it to synthetic test data of known source parameters. Pre-processing of the network input data, including the use of a rotationally invariant transform, enhanced network performance and the efficiency of the training data. The approach provides good results when applied to real magnetic data, accurately predicting the depths and layout of wall foundations within the former settlement, verified by subsequent excavation. The resulting sub-surface model is derived from the averaged outputs of a ‘committee’ of five networks, trained with individualized training sets. Fuzzy logic inference has also been used to combine individual network outputs through correlation with data from a second geophysical technique, allowing the integration of data from a separate series of measurements.

Feedforward/feedback control synthesis for performance and robustness

NASA Technical Reports Server (NTRS)

Wie, Bong; Liu, Qiang

1990-01-01

Both feedforward and feedback control approaches for uncertain dynamical systems are investigated. The control design objective is to achieve a fast settling time (high performance) and robustness (insensitivity) to plant modeling uncertainty. Preshapong of an ideal, time-optimal control input using a 'tapped-delay' filter is shown to provide a rapid maneuver with robust performance. A robust, non-minimum-phase feedback controller is synthesized with particular emphasis on its proper implementation for a non-zero set-point control problem. The proposed feedforward/feedback control approach is robust for a certain class of uncertain dynamical systems, since the control input command computed for a given desired output does not depend on the plant parameters.

Development of Feedforward Control in a Dynamic Manual Tracking Task

ERIC Educational Resources Information Center

van Roon, Dominique; Caeyenberghs, Karen; Swinnen, Stephan P.; Smits-Engelsman, Bouwien C. M.

2008-01-01

To examine the development of feedforward control during manual tracking, 117 participants in 5 age groups (6 to 7, 8 to 9, 10 to 11, 12 to 14, and 15 to 17 years) tracked an accelerating dot presented on a monitor by moving an electronic pen on a digitizer. To remain successful at higher target velocities, they had to create a predictive model of…

On the thresholds in modeling of high flows via artificial neural networks - A bootstrapping analysis

NASA Astrophysics Data System (ADS)

Panagoulia, D.; Trichakis, I.

2012-04-01

Considering the growing interest in simulating hydrological phenomena with artificial neural networks (ANNs), it is useful to figure out the potential and limits of these models. In this study, the main objective is to examine how to improve the ability of an ANN model to simulate extreme values of flow utilizing a priori knowledge of threshold values. A three-layer feedforward ANN was trained by using the back propagation algorithm and the logistic function as activation function. By using the thresholds, the flow was partitioned in low (x < μ), medium (μ ≤ x ≤ μ + 2σ) and high (x > μ + 2σ) values. The employed ANN model was trained for high flow partition and all flow data too. The developed methodology was implemented over a mountainous river catchment (the Mesochora catchment in northwestern Greece). The ANN model received as inputs pseudo-precipitation (rain plus melt) and previous observed flow data. After the training was completed the bootstrapping methodology was applied to calculate the ANN confidence intervals (CIs) for a 95% nominal coverage. The calculated CIs included only the uncertainty, which comes from the calibration procedure. The results showed that an ANN model trained specifically for high flows, with a priori knowledge of the thresholds, can simulate these extreme values much better (RMSE is 31.4% less) than an ANN model trained with all data of the available time series and using a posteriori threshold values. On the other hand the width of CIs increases by 54.9% with a simultaneous increase by 64.4% of the actual coverage for the high flows (a priori partition). The narrower CIs of the high flows trained with all data may be attributed to the smoothing effect produced from the use of the full data sets. Overall, the results suggest that an ANN model trained with a priori knowledge of the threshold values has an increased ability in simulating extreme values compared with an ANN model trained with all the data and a posteriori knowledge of the thresholds.

Speed of feedforward and recurrent processing in multilayer networks of integrate-and-fire neurons.

PubMed

Panzeri, S; Rolls, E T; Battaglia, F; Lavis, R

2001-11-01

The speed of processing in the visual cortical areas can be fast, with for example the latency of neuronal responses increasing by only approximately 10 ms per area in the ventral visual system sequence V1 to V2 to V4 to inferior temporal visual cortex. This has led to the suggestion that rapid visual processing can only be based on the feedforward connections between cortical areas. To test this idea, we investigated the dynamics of information retrieval in multiple layer networks using a four-stage feedforward network modelled with continuous dynamics with integrate-and-fire neurons, and associative synaptic connections between stages with a synaptic time constant of 10 ms. Through the implementation of continuous dynamics, we found latency differences in information retrieval of only 5 ms per layer when local excitation was absent and processing was purely feedforward. However, information latency differences increased significantly when non-associative local excitation was included. We also found that local recurrent excitation through associatively modified synapses can contribute significantly to processing in as little as 15 ms per layer, including the feedforward and local feedback processing. Moreover, and in contrast to purely feed-forward processing, the contribution of local recurrent feedback was useful and approximately this rapid even when retrieval was made difficult by noise. These findings suggest that cortical information processing can benefit from recurrent circuits when the allowed processing time per cortical area is at least 15 ms long.

Binding and segmentation via a neural mass model trained with Hebbian and anti-Hebbian mechanisms.

PubMed

Cona, Filippo; Zavaglia, Melissa; Ursino, Mauro

2012-04-01

Synchronization of neural activity in the gamma band, modulated by a slower theta rhythm, is assumed to play a significant role in binding and segmentation of multiple objects. In the present work, a recent neural mass model of a single cortical column is used to analyze the synaptic mechanisms which can warrant synchronization and desynchronization of cortical columns, during an autoassociation memory task. The model considers two distinct layers communicating via feedforward connections. The first layer receives the external input and works as an autoassociative network in the theta band, to recover a previously memorized object from incomplete information. The second realizes segmentation of different objects in the gamma band. To this end, units within both layers are connected with synapses trained on the basis of previous experience to store objects. The main model assumptions are: (i) recovery of incomplete objects is realized by excitatory synapses from pyramidal to pyramidal neurons in the same object; (ii) binding in the gamma range is realized by excitatory synapses from pyramidal neurons to fast inhibitory interneurons in the same object. These synapses (both at points i and ii) have a few ms dynamics and are trained with a Hebbian mechanism. (iii) Segmentation is realized with faster AMPA synapses, with rise times smaller than 1 ms, trained with an anti-Hebbian mechanism. Results show that the model, with the previous assumptions, can correctly reconstruct and segment three simultaneous objects, starting from incomplete knowledge. Segmentation of more objects is possible but requires an increased ratio between the theta and gamma periods.

On the use of ANN interconnection weights in optimal structural design

NASA Technical Reports Server (NTRS)

Hajela, P.; Szewczyk, Z.

1992-01-01

The present paper describes the use of interconnection weights of a multilayer, feedforward network, to extract information pertinent to the mapping space that the network is assumed to represent. In particular, these weights can be used to determine an appropriate network architecture, and an adequate number of training patterns (input-output pairs) have been used for network training. The weight analysis also provides an approach to assess the influence of each input parameter on a selected output component. The paper shows the significance of this information in decomposition driven optimal design.

Neural network-based run-to-run controller using exposure and resist thickness adjustment

NASA Astrophysics Data System (ADS)

Geary, Shane; Barry, Ronan

2003-06-01

This paper describes the development of a run-to-run control algorithm using a feedforward neural network, trained using the backpropagation training method. The algorithm is used to predict the critical dimension of the next lot using previous lot information. It is compared to a common prediction algorithm - the exponentially weighted moving average (EWMA) and is shown to give superior prediction performance in simulations. The manufacturing implementation of the final neural network showed significantly improved process capability when compared to the case where no run-to-run control was utilised.

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.

A multifactorial analysis of obesity as CVD risk factor: Use of neural network based methods in a nutrigenetics context

PubMed Central

2010-01-01

Background Obesity is a multifactorial trait, which comprises an independent risk factor for cardiovascular disease (CVD). The aim of the current work is to study the complex etiology beneath obesity and identify genetic variations and/or factors related to nutrition that contribute to its variability. To this end, a set of more than 2300 white subjects who participated in a nutrigenetics study was used. For each subject a total of 63 factors describing genetic variants related to CVD (24 in total), gender, and nutrition (38 in total), e.g. average daily intake in calories and cholesterol, were measured. Each subject was categorized according to body mass index (BMI) as normal (BMI ≤ 25) or overweight (BMI > 25). Two artificial neural network (ANN) based methods were designed and used towards the analysis of the available data. These corresponded to i) a multi-layer feed-forward ANN combined with a parameter decreasing method (PDM-ANN), and ii) a multi-layer feed-forward ANN trained by a hybrid method (GA-ANN) which combines genetic algorithms and the popular back-propagation training algorithm. Results PDM-ANN and GA-ANN were comparatively assessed in terms of their ability to identify the most important factors among the initial 63 variables describing genetic variations, nutrition and gender, able to classify a subject into one of the BMI related classes: normal and overweight. The methods were designed and evaluated using appropriate training and testing sets provided by 3-fold Cross Validation (3-CV) resampling. Classification accuracy, sensitivity, specificity and area under receiver operating characteristics curve were utilized to evaluate the resulted predictive ANN models. The most parsimonious set of factors was obtained by the GA-ANN method and included gender, six genetic variations and 18 nutrition-related variables. The corresponding predictive model was characterized by a mean accuracy equal of 61.46% in the 3-CV testing sets. Conclusions The ANN based methods revealed factors that interactively contribute to obesity trait and provided predictive models with a promising generalization ability. In general, results showed that ANNs and their hybrids can provide useful tools for the study of complex traits in the context of nutrigenetics. PMID:20825661

Recurrent Network models of sequence generation and memory

PubMed Central

Rajan, Kanaka; Harvey, Christopher D; Tank, David W

2016-01-01

SUMMARY Sequential activation of neurons is a common feature of network activity during a variety of behaviors, including working memory and decision making. Previous network models for sequences and memory emphasized specialized architectures in which a principled mechanism is pre-wired into their connectivity. Here, we demonstrate that starting from random connectivity and modifying a small fraction of connections, a largely disordered recurrent network can produce sequences and implement working memory efficiently. We use this process, called Partial In-Network training (PINning), to model and match cellular-resolution imaging data from the posterior parietal cortex during a virtual memory-guided two-alternative forced choice task [Harvey, Coen and Tank, 2012]. Analysis of the connectivity reveals that sequences propagate by the cooperation between recurrent synaptic interactions and external inputs, rather than through feedforward or asymmetric connections. Together our results suggest that neural sequences may emerge through learning from largely unstructured network architectures. PMID:26971945

A Spiking Neural Network Based Cortex-Like Mechanism and Application to Facial Expression Recognition

PubMed Central

Fu, Si-Yao; Yang, Guo-Sheng; Kuai, Xin-Kai

2012-01-01

In this paper, we present a quantitative, highly structured cortex-simulated model, which can be simply described as feedforward, hierarchical simulation of ventral stream of visual cortex using biologically plausible, computationally convenient spiking neural network system. The motivation comes directly from recent pioneering works on detailed functional decomposition analysis of the feedforward pathway of the ventral stream of visual cortex and developments on artificial spiking neural networks (SNNs). By combining the logical structure of the cortical hierarchy and computing power of the spiking neuron model, a practical framework has been presented. As a proof of principle, we demonstrate our system on several facial expression recognition tasks. The proposed cortical-like feedforward hierarchy framework has the merit of capability of dealing with complicated pattern recognition problems, suggesting that, by combining the cognitive models with modern neurocomputational approaches, the neurosystematic approach to the study of cortex-like mechanism has the potential to extend our knowledge of brain mechanisms underlying the cognitive analysis and to advance theoretical models of how we recognize face or, more specifically, perceive other people's facial expression in a rich, dynamic, and complex environment, providing a new starting point for improved models of visual cortex-like mechanism. PMID:23193391

A spiking neural network based cortex-like mechanism and application to facial expression recognition.

PubMed

Fu, Si-Yao; Yang, Guo-Sheng; Kuai, Xin-Kai

2012-01-01

In this paper, we present a quantitative, highly structured cortex-simulated model, which can be simply described as feedforward, hierarchical simulation of ventral stream of visual cortex using biologically plausible, computationally convenient spiking neural network system. The motivation comes directly from recent pioneering works on detailed functional decomposition analysis of the feedforward pathway of the ventral stream of visual cortex and developments on artificial spiking neural networks (SNNs). By combining the logical structure of the cortical hierarchy and computing power of the spiking neuron model, a practical framework has been presented. As a proof of principle, we demonstrate our system on several facial expression recognition tasks. The proposed cortical-like feedforward hierarchy framework has the merit of capability of dealing with complicated pattern recognition problems, suggesting that, by combining the cognitive models with modern neurocomputational approaches, the neurosystematic approach to the study of cortex-like mechanism has the potential to extend our knowledge of brain mechanisms underlying the cognitive analysis and to advance theoretical models of how we recognize face or, more specifically, perceive other people's facial expression in a rich, dynamic, and complex environment, providing a new starting point for improved models of visual cortex-like mechanism.

An application of artificial neural networks to experimental data approximation

NASA Technical Reports Server (NTRS)

Meade, Andrew J., Jr.

1993-01-01

As an initial step in the evaluation of networks, a feedforward architecture is trained to approximate experimental data by the backpropagation algorithm. Several drawbacks were detected and an alternative learning algorithm was then developed to partially address the drawbacks. This noniterative algorithm has a number of advantages over the backpropagation method and is easily implemented on existing hardware.

Dispersion compensation of fiber optic communication system with direct detection using artificial neural networks (ANNs)

NASA Astrophysics Data System (ADS)

Maghrabi, Mahmoud M. T.; Kumar, Shiva; Bakr, Mohamed H.

2018-02-01

This work introduces a powerful digital nonlinear feed-forward equalizer (NFFE), exploiting multilayer artificial neural network (ANN). It mitigates impairments of optical communication systems arising due to the nonlinearity introduced by direct photo-detection. In a direct detection system, the detection process is nonlinear due to the fact that the photo-current is proportional to the absolute square of the electric field intensity. The proposed equalizer provides the most efficient computational cost with high equalization performance. Its performance is comparable to the benchmark compensation performance achieved by maximum-likelihood sequence estimator. The equalizer trains an ANN to act as a nonlinear filter whose impulse response removes the intersymbol interference (ISI) distortions of the optical channel. Owing to the proposed extensive training of the equalizer, it achieves the ultimate performance limit of any feed-forward equalizer (FFE). The performance and efficiency of the equalizer is investigated by applying it to various practical short-reach fiber optic communication system scenarios. These scenarios are extracted from practical metro/media access networks and data center applications. The obtained results show that the ANN-NFFE compensates for the received BER degradation and significantly increases the tolerance to the chromatic dispersion distortion.

Improving Classification Performance through an Advanced Ensemble Based Heterogeneous Extreme Learning Machines.

PubMed

Abuassba, Adnan O M; Zhang, Dezheng; Luo, Xiong; Shaheryar, Ahmad; Ali, Hazrat

2017-01-01

Extreme Learning Machine (ELM) is a fast-learning algorithm for a single-hidden layer feedforward neural network (SLFN). It often has good generalization performance. However, there are chances that it might overfit the training data due to having more hidden nodes than needed. To address the generalization performance, we use a heterogeneous ensemble approach. We propose an Advanced ELM Ensemble (AELME) for classification, which includes Regularized-ELM, L 2 -norm-optimized ELM (ELML2), and Kernel-ELM. The ensemble is constructed by training a randomly chosen ELM classifier on a subset of training data selected through random resampling. The proposed AELM-Ensemble is evolved by employing an objective function of increasing diversity and accuracy among the final ensemble. Finally, the class label of unseen data is predicted using majority vote approach. Splitting the training data into subsets and incorporation of heterogeneous ELM classifiers result in higher prediction accuracy, better generalization, and a lower number of base classifiers, as compared to other models (Adaboost, Bagging, Dynamic ELM ensemble, data splitting ELM ensemble, and ELM ensemble). The validity of AELME is confirmed through classification on several real-world benchmark datasets.

Improving Classification Performance through an Advanced Ensemble Based Heterogeneous Extreme Learning Machines

PubMed Central

Abuassba, Adnan O. M.; Ali, Hazrat

2017-01-01

Extreme Learning Machine (ELM) is a fast-learning algorithm for a single-hidden layer feedforward neural network (SLFN). It often has good generalization performance. However, there are chances that it might overfit the training data due to having more hidden nodes than needed. To address the generalization performance, we use a heterogeneous ensemble approach. We propose an Advanced ELM Ensemble (AELME) for classification, which includes Regularized-ELM, L2-norm-optimized ELM (ELML2), and Kernel-ELM. The ensemble is constructed by training a randomly chosen ELM classifier on a subset of training data selected through random resampling. The proposed AELM-Ensemble is evolved by employing an objective function of increasing diversity and accuracy among the final ensemble. Finally, the class label of unseen data is predicted using majority vote approach. Splitting the training data into subsets and incorporation of heterogeneous ELM classifiers result in higher prediction accuracy, better generalization, and a lower number of base classifiers, as compared to other models (Adaboost, Bagging, Dynamic ELM ensemble, data splitting ELM ensemble, and ELM ensemble). The validity of AELME is confirmed through classification on several real-world benchmark datasets. PMID:28546808

Generalised Transfer Functions of Neural Networks

NASA Astrophysics Data System (ADS)

Fung, C. F.; Billings, S. A.; Zhang, H.

1997-11-01

When artificial neural networks are used to model non-linear dynamical systems, the system structure which can be extremely useful for analysis and design, is buried within the network architecture. In this paper, explicit expressions for the frequency response or generalised transfer functions of both feedforward and recurrent neural networks are derived in terms of the network weights. The derivation of the algorithm is established on the basis of the Taylor series expansion of the activation functions used in a particular neural network. This leads to a representation which is equivalent to the non-linear recursive polynomial model and enables the derivation of the transfer functions to be based on the harmonic expansion method. By mapping the neural network into the frequency domain information about the structure of the underlying non-linear system can be recovered. Numerical examples are included to demonstrate the application of the new algorithm. These examples show that the frequency response functions appear to be highly sensitive to the network topology and training, and that the time domain properties fail to reveal deficiencies in the trained network structure.

Neural substrates of visuomotor learning based on improved feedback control and prediction

PubMed Central

Grafton, Scott T.; Schmitt, Paul; Horn, John Van; Diedrichsen, Jörn

2008-01-01

Motor skills emerge from learning feedforward commands as well as improvements in feedback control. These two components of learning were investigated in a compensatory visuomotor tracking task on a trial-by-trial basis. Between trial learning was characterized with a state-space model to provide smoothed estimates of feedforward and feedback learning, separable from random fluctuations in motor performance and error. The resultant parameters were correlated with brain activity using magnetic resonance imaging. Learning related to the generation of a feedforward command correlated with activity in dorsal premotor cortex, inferior parietal lobule, supplementary motor area and cingulate motor area, supporting a role of these areas in retrieving and executing a predictive motor command. Modulation of feedback control was associated with activity in bilateral posterior superior parietal lobule as well as right ventral premotor cortex. Performance error correlated with activity in a widespread cortical and subcortical network including bilateral parietal, premotor and rostral anterior cingulate cortex as well as the cerebellar cortex. Finally, trial-by-trial changes of kinematics, as measured by mean absolute hand acceleration, correlated with activity in motor cortex and anterior cerebellum. The results demonstrate that incremental, learning dependent changes can be modeled on a trial-by-trial basis and neural substrates for feedforward control of novel motor programs are localized to secondary motor areas. PMID:18032069

Propagating synchrony in feed-forward networks

PubMed Central

Jahnke, Sven; Memmesheimer, Raoul-Martin; Timme, Marc

2013-01-01

Coordinated patterns of precisely timed action potentials (spikes) emerge in a variety of neural circuits but their dynamical origin is still not well understood. One hypothesis states that synchronous activity propagating through feed-forward chains of groups of neurons (synfire chains) may dynamically generate such spike patterns. Additionally, synfire chains offer the possibility to enable reliable signal transmission. So far, mostly densely connected chains, often with all-to-all connectivity between groups, have been theoretically and computationally studied. Yet, such prominent feed-forward structures have not been observed experimentally. Here we analytically and numerically investigate under which conditions diluted feed-forward chains may exhibit synchrony propagation. In addition to conventional linear input summation, we study the impact of non-linear, non-additive summation accounting for the effect of fast dendritic spikes. The non-linearities promote synchronous inputs to generate precisely timed spikes. We identify how non-additive coupling relaxes the conditions on connectivity such that it enables synchrony propagation at connectivities substantially lower than required for linearly coupled chains. Although the analytical treatment is based on a simple leaky integrate-and-fire neuron model, we show how to generalize our methods to biologically more detailed neuron models and verify our results by numerical simulations with, e.g., Hodgkin Huxley type neurons. PMID:24298251

Modeling the Time-Course of Responses for the Border Ownership Selectivity Based on the Integration of Feedforward Signals and Visual Cortical Interactions

PubMed Central

Wagatsuma, Nobuhiko; Sakai, Ko

2017-01-01

Border ownership (BO) indicates which side of a contour owns a border, and it plays a fundamental role in figure-ground segregation. The majority of neurons in V2 and V4 areas of monkeys exhibit BO selectivity. A physiological work reported that the responses of BO-selective cells show a rapid transition when a presented square is flipped along its classical receptive field (CRF) so that the opposite BO is presented, whereas the transition is significantly slower when a square with a clear BO is replaced by an ambiguous edge, e.g., when the square is enlarged greatly. The rapid transition seemed to reflect the influence of feedforward processing on BO selectivity. Herein, we investigated the role of feedforward signals and cortical interactions for time-courses in BO-selective cells by modeling a visual cortical network comprising V1, V2, and posterior parietal (PP) modules. In our computational model, the recurrent pathways among these modules gradually established the visual progress and the BO assignments. Feedforward inputs mainly determined the activities of these modules. Surrounding suppression/facilitation of early-level areas modulates the activities of V2 cells to provide BO signals. Weak feedback signals from the PP module enhanced the contrast gain extracted in V1, which underlies the attentional modulation of BO signals. Model simulations exhibited time-courses depending on the BO ambiguity, which were caused by the integration delay of V1 and V2 cells and the local inhibition therein given the difference in input stimulus. However, our model did not fully explain the characteristics of crucially slow transition: the responses of BO-selective physiological cells indicated the persistent activation several times longer than that of our model after the replacement with the ambiguous edge. Furthermore, the time-course of BO-selective model cells replicated the attentional modulation of response time in human psychophysical experiments. These attentional modulations for time-courses were induced by selective enhancement of early-level features due to interactions between V1 and PP. Our proposed model suggests fundamental roles of surrounding suppression/facilitation based on feedforward inputs as well as the interactions between early and parietal visual areas with respect to the ambiguity dependence of the neural dynamics in intermediate-level vision. PMID:28163688

Modeling the Time-Course of Responses for the Border Ownership Selectivity Based on the Integration of Feedforward Signals and Visual Cortical Interactions.

PubMed

Wagatsuma, Nobuhiko; Sakai, Ko

2016-01-01

Border ownership (BO) indicates which side of a contour owns a border, and it plays a fundamental role in figure-ground segregation. The majority of neurons in V2 and V4 areas of monkeys exhibit BO selectivity. A physiological work reported that the responses of BO-selective cells show a rapid transition when a presented square is flipped along its classical receptive field (CRF) so that the opposite BO is presented, whereas the transition is significantly slower when a square with a clear BO is replaced by an ambiguous edge, e.g., when the square is enlarged greatly. The rapid transition seemed to reflect the influence of feedforward processing on BO selectivity. Herein, we investigated the role of feedforward signals and cortical interactions for time-courses in BO-selective cells by modeling a visual cortical network comprising V1, V2, and posterior parietal (PP) modules. In our computational model, the recurrent pathways among these modules gradually established the visual progress and the BO assignments. Feedforward inputs mainly determined the activities of these modules. Surrounding suppression/facilitation of early-level areas modulates the activities of V2 cells to provide BO signals. Weak feedback signals from the PP module enhanced the contrast gain extracted in V1, which underlies the attentional modulation of BO signals. Model simulations exhibited time-courses depending on the BO ambiguity, which were caused by the integration delay of V1 and V2 cells and the local inhibition therein given the difference in input stimulus. However, our model did not fully explain the characteristics of crucially slow transition: the responses of BO-selective physiological cells indicated the persistent activation several times longer than that of our model after the replacement with the ambiguous edge. Furthermore, the time-course of BO-selective model cells replicated the attentional modulation of response time in human psychophysical experiments. These attentional modulations for time-courses were induced by selective enhancement of early-level features due to interactions between V1 and PP. Our proposed model suggests fundamental roles of surrounding suppression/facilitation based on feedforward inputs as well as the interactions between early and parietal visual areas with respect to the ambiguity dependence of the neural dynamics in intermediate-level vision.

Subtractive, divisive and non-monotonic gain control in feedforward nets linearized by noise and delays.

PubMed

Mejias, Jorge F; Payeur, Alexandre; Selin, Erik; Maler, Leonard; Longtin, André

2014-01-01

The control of input-to-output mappings, or gain control, is one of the main strategies used by neural networks for the processing and gating of information. Using a spiking neural network model, we studied the gain control induced by a form of inhibitory feedforward circuitry-also known as "open-loop feedback"-, which has been experimentally observed in a cerebellum-like structure in weakly electric fish. We found, both analytically and numerically, that this network displays three different regimes of gain control: subtractive, divisive, and non-monotonic. Subtractive gain control was obtained when noise is very low in the network. Also, it was possible to change from divisive to non-monotonic gain control by simply modulating the strength of the feedforward inhibition, which may be achieved via long-term synaptic plasticity. The particular case of divisive gain control has been previously observed in vivo in weakly electric fish. These gain control regimes were robust to the presence of temporal delays in the inhibitory feedforward pathway, which were found to linearize the input-to-output mappings (or f-I curves) via a novel variability-increasing mechanism. Our findings highlight the feedforward-induced gain control analyzed here as a highly versatile mechanism of information gating in the brain.

Subtractive, divisive and non-monotonic gain control in feedforward nets linearized by noise and delays

PubMed Central

Mejias, Jorge F.; Payeur, Alexandre; Selin, Erik; Maler, Leonard; Longtin, André

2014-01-01

The control of input-to-output mappings, or gain control, is one of the main strategies used by neural networks for the processing and gating of information. Using a spiking neural network model, we studied the gain control induced by a form of inhibitory feedforward circuitry—also known as “open-loop feedback”—, which has been experimentally observed in a cerebellum-like structure in weakly electric fish. We found, both analytically and numerically, that this network displays three different regimes of gain control: subtractive, divisive, and non-monotonic. Subtractive gain control was obtained when noise is very low in the network. Also, it was possible to change from divisive to non-monotonic gain control by simply modulating the strength of the feedforward inhibition, which may be achieved via long-term synaptic plasticity. The particular case of divisive gain control has been previously observed in vivo in weakly electric fish. These gain control regimes were robust to the presence of temporal delays in the inhibitory feedforward pathway, which were found to linearize the input-to-output mappings (or f-I curves) via a novel variability-increasing mechanism. Our findings highlight the feedforward-induced gain control analyzed here as a highly versatile mechanism of information gating in the brain. PMID:24616694

Geometry-based across wafer process control in a dual damascene scenario

NASA Astrophysics Data System (ADS)

Krause, Gerd; Hofmann, Detlef; Habets, Boris; Buhl, Stefan; Gutsch, Manuela; Lopez-Gomez, Alberto; Thrun, Xaver

2018-03-01

Dual damascene is an established patterning process for back-end-of-line to generate copper interconnects and lines. One of the critical output parameters is the electrical resistance of the metal lines. In our 200 mm line, this is currently being controlled by a feed-forward control from the etch process to the final step in the CMP process. In this paper, we investigate the impact of alternative feed-forward control using a calibrated physical model that estimates the impact on electrical resistance of the metal lines* . This is done by simulation on a large set of wafers. Three different approaches are evaluated, one of which uses different feed-forward settings for different radial zones in the CMP process.

Adaptive filter design using recurrent cerebellar model articulation controller.

PubMed

Lin, Chih-Min; Chen, Li-Yang; Yeung, Daniel S

2010-07-01

A novel adaptive filter is proposed using a recurrent cerebellar-model-articulation-controller (CMAC). The proposed locally recurrent globally feedforward recurrent CMAC (RCMAC) has favorable properties of small size, good generalization, rapid learning, and dynamic response, thus it is more suitable for high-speed signal processing. To provide fast training, an efficient parameter learning algorithm based on the normalized gradient descent method is presented, in which the learning rates are on-line adapted. Then the Lyapunov function is utilized to derive the conditions of the adaptive learning rates, so the stability of the filtering error can be guaranteed. To demonstrate the performance of the proposed adaptive RCMAC filter, it is applied to a nonlinear channel equalization system and an adaptive noise cancelation system. The advantages of the proposed filter over other adaptive filters are verified through simulations.

A Novel Higher Order Artificial Neural Networks

NASA Astrophysics Data System (ADS)

Xu, Shuxiang

2010-05-01

In this paper a new Higher Order Neural Network (HONN) model is introduced and applied in several data mining tasks. Data Mining extracts hidden patterns and valuable information from large databases. A hyperbolic tangent function is used as the neuron activation function for the new HONN model. Experiments are conducted to demonstrate the advantages and disadvantages of the new HONN model, when compared with several conventional Artificial Neural Network (ANN) models: Feedforward ANN with the sigmoid activation function; Feedforward ANN with the hyperbolic tangent activation function; and Radial Basis Function (RBF) ANN with the Gaussian activation function. The experimental results seem to suggest that the new HONN holds higher generalization capability as well as abilities in handling missing data.

Integration of Online Parameter Identification and Neural Network for In-Flight Adaptive Control

NASA Technical Reports Server (NTRS)

Hageman, Jacob J.; Smith, Mark S.; Stachowiak, Susan

2003-01-01

An indirect adaptive system has been constructed for robust control of an aircraft with uncertain aerodynamic characteristics. This system consists of a multilayer perceptron pre-trained neural network, online stability and control derivative identification, a dynamic cell structure online learning neural network, and a model following control system based on the stochastic optimal feedforward and feedback technique. The pre-trained neural network and model following control system have been flight-tested, but the online parameter identification and online learning neural network are new additions used for in-flight adaptation of the control system model. A description of the modification and integration of these two stand-alone software packages into the complete system in preparation for initial flight tests is presented. Open-loop results using both simulation and flight data, as well as closed-loop performance of the complete system in a nonlinear, six-degree-of-freedom, flight validated simulation, are analyzed. Results show that this online learning system, in contrast to the nonlearning system, has the ability to adapt to changes in aerodynamic characteristics in a real-time, closed-loop, piloted simulation, resulting in improved flying qualities.

Biological engineering applications of feedforward neural networks designed and parameterized by genetic algorithms.

PubMed

Ferentinos, Konstantinos P

2005-09-01

Two neural network (NN) applications in the field of biological engineering are developed, designed and parameterized by an evolutionary method based on the evolutionary process of genetic algorithms. The developed systems are a fault detection NN model and a predictive modeling NN system. An indirect or 'weak specification' representation was used for the encoding of NN topologies and training parameters into genes of the genetic algorithm (GA). Some a priori knowledge of the demands in network topology for specific application cases is required by this approach, so that the infinite search space of the problem is limited to some reasonable degree. Both one-hidden-layer and two-hidden-layer network architectures were explored by the GA. Except for the network architecture, each gene of the GA also encoded the type of activation functions in both hidden and output nodes of the NN and the type of minimization algorithm that was used by the backpropagation algorithm for the training of the NN. Both models achieved satisfactory performance, while the GA system proved to be a powerful tool that can successfully replace the problematic trial-and-error approach that is usually used for these tasks.

Using an electrohydraulic ankle foot orthosis to study modifications in feedforward control during locomotor adaptation to force fields applied in stance

PubMed Central

Noel, Martin; Fortin, Karine; Bouyer, Laurent J

2009-01-01

Background Adapting to external forces during walking has been proposed as a tool to improve locomotion after central nervous system injury. However, sensorimotor integration during walking varies according to the timing in the gait cycle, suggesting that adaptation may also depend on gait phases. In this study, an ElectroHydraulic AFO (EHO) was used to apply forces specifically during mid-stance and push-off to evaluate if feedforward movement control can be adapted in these 2 gait phases. Methods Eleven healthy subjects walked on a treadmill before (3 min), during (5 min) and after (5 min) exposure to 2 force fields applied by the EHO (mid-stance/push-off; ~10 Nm, towards dorsiflexion). To evaluate modifications in feedforward control, strides with no force field ('catch strides') were unexpectedly inserted during the force field walking period. Results When initially exposed to a mid-stance force field (FF20%), subjects showed a significant increase in ankle dorsiflexion velocity. Catches applied early into the FF20% were similar to baseline (P > 0.99). Subjects gradually adapted by returning ankle velocity to baseline over ~50 strides. Catches applied thereafter showed decreased ankle velocity where the force field was normally applied, indicating the presence of feedforward adaptation. When initially exposed to a push-off force field (FF50%), plantarflexion velocity was reduced in the zone of force field application. No adaptation occurred over the 5 min exposure. Catch strides kinematics remained similar to control at all times, suggesting no feedforward adaptation. As a control, force fields assisting plantarflexion (-3.5 to -9.5 Nm) were applied and increased ankle plantarflexion during push-off, confirming that the lack of kinematic changes during FF50% catch strides were not simply due to a large ankle impedance. Conclusion Together these results show that ankle exoskeletons such as the EHO can be used to study phase-specific adaptive control of the ankle during locomotion. Our data suggest that, for short duration exposure, a feedforward modification in torque output occurs during mid-stance but not during push-off. These findings are important for the design of novel rehabilitation methods, as they suggest that the ability to use resistive force fields for training may depend on targeted gait phases. PMID:19493356

Using an electrohydraulic ankle foot orthosis to study modifications in feedforward control during locomotor adaptation to force fields applied in stance.

PubMed

Noel, Martin; Fortin, Karine; Bouyer, Laurent J

2009-06-03

Adapting to external forces during walking has been proposed as a tool to improve locomotion after central nervous system injury. However, sensorimotor integration during walking varies according to the timing in the gait cycle, suggesting that adaptation may also depend on gait phases. In this study, an ElectroHydraulic AFO (EHO) was used to apply forces specifically during mid-stance and push-off to evaluate if feedforward movement control can be adapted in these 2 gait phases. Eleven healthy subjects walked on a treadmill before (3 min), during (5 min) and after (5 min) exposure to 2 force fields applied by the EHO (mid-stance/push-off; approximately 10 Nm, towards dorsiflexion). To evaluate modifications in feedforward control, strides with no force field ('catch strides') were unexpectedly inserted during the force field walking period. When initially exposed to a mid-stance force field (FF 20%), subjects showed a significant increase in ankle dorsiflexion velocity. Catches applied early into the FF 20% were similar to baseline (P > 0.99). Subjects gradually adapted by returning ankle velocity to baseline over approximately 50 strides. Catches applied thereafter showed decreased ankle velocity where the force field was normally applied, indicating the presence of feedforward adaptation. When initially exposed to a push-off force field (FF 50%), plantarflexion velocity was reduced in the zone of force field application. No adaptation occurred over the 5 min exposure. Catch strides kinematics remained similar to control at all times, suggesting no feedforward adaptation. As a control, force fields assisting plantarflexion (-3.5 to -9.5 Nm) were applied and increased ankle plantarflexion during push-off, confirming that the lack of kinematic changes during FF 50% catch strides were not simply due to a large ankle impedance. Together these results show that ankle exoskeletons such as the EHO can be used to study phase-specific adaptive control of the ankle during locomotion. Our data suggest that, for short duration exposure, a feedforward modification in torque output occurs during mid-stance but not during push-off. These findings are important for the design of novel rehabilitation methods, as they suggest that the ability to use resistive force fields for training may depend on targeted gait phases.

Parallel approach to identifying the well-test interpretation model using a neurocomputer

NASA Astrophysics Data System (ADS)

May, Edward A., Jr.; Dagli, Cihan H.

1996-03-01

The well test is one of the primary diagnostic and predictive tools used in the analysis of oil and gas wells. In these tests, a pressure recording device is placed in the well and the pressure response is recorded over time under controlled flow conditions. The interpreted results are indicators of the well's ability to flow and the damage done to the formation surrounding the wellbore during drilling and completion. The results are used for many purposes, including reservoir modeling (simulation) and economic forecasting. The first step in the analysis is the identification of the Well-Test Interpretation (WTI) model, which determines the appropriate solution method. Mis-identification of the WTI model occurs due to noise and non-ideal reservoir conditions. Previous studies have shown that a feed-forward neural network using the backpropagation algorithm can be used to identify the WTI model. One of the drawbacks to this approach is, however, training time, which can run into days of CPU time on personal computers. In this paper a similar neural network is applied using both a personal computer and a neurocomputer. Input data processing, network design, and performance are discussed and compared. The results show that the neurocomputer greatly eases the burden of training and allows the network to outperform a similar network running on a personal computer.

Encoding Time in Feedforward Trajectories of a Recurrent Neural Network Model.

PubMed

Hardy, N F; Buonomano, Dean V

2018-02-01

Brain activity evolves through time, creating trajectories of activity that underlie sensorimotor processing, behavior, and learning and memory. Therefore, understanding the temporal nature of neural dynamics is essential to understanding brain function and behavior. In vivo studies have demonstrated that sequential transient activation of neurons can encode time. However, it remains unclear whether these patterns emerge from feedforward network architectures or from recurrent networks and, furthermore, what role network structure plays in timing. We address these issues using a recurrent neural network (RNN) model with distinct populations of excitatory and inhibitory units. Consistent with experimental data, a single RNN could autonomously produce multiple functionally feedforward trajectories, thus potentially encoding multiple timed motor patterns lasting up to several seconds. Importantly, the model accounted for Weber's law, a hallmark of timing behavior. Analysis of network connectivity revealed that efficiency-a measure of network interconnectedness-decreased as the number of stored trajectories increased. Additionally, the balance of excitation (E) and inhibition (I) shifted toward excitation during each unit's activation time, generating the prediction that observed sequential activity relies on dynamic control of the E/I balance. Our results establish for the first time that the same RNN can generate multiple functionally feedforward patterns of activity as a result of dynamic shifts in the E/I balance imposed by the connectome of the RNN. We conclude that recurrent network architectures account for sequential neural activity, as well as for a fundamental signature of timing behavior: Weber's law.

Simulated mossy fiber associated feedforward circuit functioning as a highpass filter.

PubMed

Zalay, Osbert C; Bardakjian, Berj L

2006-01-01

Learning and memory rely on the strict regulation of communication between neurons in the hippocampus. The mossy fiber (MF) pathway connects the dentate gyrus to the auto-associative CA3 network, and the information it carries is controlled by a feedforward circuit combining disynaptic inhibition with monosynaptic excitation. Analysis of the MF associated circuit using a mapped clock oscillator (MCO) model reveals the circuit to be a highpass filter.

Comparison between hybrid feedforward-feedback, feedforward, and feedback structures for active noise control of fMRI noise.

PubMed

Reddy, Rajiv M; Panahi, Issa M S

2008-01-01

The performance of FIR feedforward, IIR feedforward, FIR feedback, hybrid FIR feedforward--FIR feedback, and hybrid IIR feedforward - FIR feedback structures for active noise control (ANC) are compared for an fMRI noise application. The filtered-input normalized least squares (FxNLMS) algorithm is used to update the coefficients of the adaptive filters in all these structures. Realistic primary and secondary paths of an fMRI bore are used by estimating them on a half cylindrical acrylic bore of 0.76 m (D)x1.52 m (L). Detailed results of the performance of the ANC system are presented in the paper for each of these structures. We find that the IIR feedforward structure produces most of the performance improvement in the hybrid IIR feedforward - FIR feedback structure and adding the feedback structure becomes almost redundant in the case of fMRI noise.

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.

Development of programmable artificial neural networks

NASA Technical Reports Server (NTRS)

Meade, Andrew J.

1993-01-01

Conventionally programmed digital computers can process numbers with great speed and precision, but do not easily recognize patterns or imprecise or contradictory data. Instead of being programmed in the conventional sense, artificial neural networks are capable of self-learning through exposure to repeated examples. However, the training of an ANN can be a time consuming and unpredictable process. A general method is being developed to mate the adaptability of the ANN with the speed and precision of the digital computer. This method was successful in building feedforward networks that can approximate functions and their partial derivatives from examples in a single iteration. The general method also allows the formation of feedforward networks that can approximate the solution to nonlinear ordinary and partial differential equations to desired accuracy without the need of examples. It is believed that continued research will produce artificial neural networks that can be used with confidence in practical scientific computing and engineering applications.

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 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.

Modeling of feed-forward control using the partial least squares regression method in the tablet compression process.

PubMed

Hattori, Yusuke; Otsuka, Makoto

2017-05-30

In the pharmaceutical industry, the implementation of continuous manufacturing has been widely promoted in lieu of the traditional batch manufacturing approach. More specially, in recent years, the innovative concept of feed-forward control has been introduced in relation to process analytical technology. In the present study, we successfully developed a feed-forward control model for the tablet compression process by integrating data obtained from near-infrared (NIR) spectra and the physical properties of granules. In the pharmaceutical industry, batch manufacturing routinely allows for the preparation of granules with the desired properties through the manual control of process parameters. On the other hand, continuous manufacturing demands the automatic determination of these process parameters. Here, we proposed the development of a control model using the partial least squares regression (PLSR) method. The most significant feature of this method is the use of dataset integrating both the NIR spectra and the physical properties of the granules. Using our model, we determined that the properties of products, such as tablet weight and thickness, need to be included as independent variables in the PLSR analysis in order to predict unknown process parameters. Copyright © 2017 Elsevier B.V. All rights reserved.

Analysis of bidirectional pattern synchrony of concentration-secretion pairs: implementation in the human testicular and adrenal axes.

PubMed

Liu, Peter Y; Pincus, Steven M; Keenan, Daniel M; Roelfsema, Ferdinand; Veldhuis, Johannes D

2005-02-01

The hypothalamo-pituitary-testicular and hypothalamo-pituitary-adrenal axes are prototypical coupled neuroendocrine systems. In the present study, we contrasted in vivo linkages within and between these two axes using methods without linearity assumptions. We examined 11 young (21-31 yr) and 8 older (62-74 yr) men who underwent frequent (every 2.5 min) blood sampling overnight for paired measurement of LH and testosterone and 35 adults (17 women and 18 men; 26-77 yr old) who underwent adrenocorticotropic hormone (ACTH) and cortisol measurements every 10 min for 24 h. To mirror physiological interactions, hormone secretion was first deconvolved from serial concentrations with a waveform-independent biexponential elimination model. Feedforward synchrony, feedback synchrony, and the difference in feedforward-feedback synchrony were quantified by the cross-approximate entropy (X-ApEn) statistic. These were applied in a forward (LH concentration template, examining pattern recurrence in testosterone secretion), reverse (testosterone concentration template, examining pattern recurrence in LH secretion), and differential (forward minus reverse) manner, respectively. Analogous concentration-secretion X-ApEn estimates were calculated from ACTH-cortisol pairs. X-ApEn, a scale- and model-independent measure of pattern reproducibility, disclosed 1) greater testosterone-LH feedback coordination than LH-testosterone feedforward synchrony in healthy men and significant and symmetric erosion of both feedforward and feedback linkages with aging; 2) more synchronous ACTH concentration-dependent feedforward than feedback drive of cortisol secretion, independent of gender and age; and 3) enhanced detection of bidirectional physiological regulation by in vivo pairwise concentration-secretion compared with concentration-concentration analyses. The linking of relevant biological input to output signals and vice versa should be useful in the dissection of the reciprocal control of neuroendocrine systems or even in the analysis of other nonendocrine networks.

Least square neural network model of the crude oil blending process.

PubMed

Rubio, José de Jesús

2016-06-01

In this paper, the recursive least square algorithm is designed for the big data learning of a feedforward neural network. The proposed method as the combination of the recursive least square and feedforward neural network obtains four advantages over the alone algorithms: it requires less number of regressors, it is fast, it has the learning ability, and it is more compact. Stability, convergence, boundedness of parameters, and local minimum avoidance of the proposed technique are guaranteed. The introduced strategy is applied for the modeling of the crude oil blending process. Copyright © 2016 Elsevier Ltd. All rights reserved.

Combustion Control System Design of Diesel Engine via ASPR based Output Feedback Control Strategy with a PFC

NASA Astrophysics Data System (ADS)

Mizumoto, Ikuro; Tsunematsu, Junpei; Fujii, Seiya

2016-09-01

In this paper, a design method of an output feedback control system with a simple feedforward input for a combustion model of diesel engine will be proposed based on the almost strictly positive real-ness (ASPR-ness) of the controlled system for a combustion control of diesel engines. A parallel feedforward compensator (PFC) design scheme which renders the resulting augmented controlled system ASPR will also be proposed in order to design a stable output feedback control system for the considered combustion model. The effectiveness of our proposed method will be confirmed through numerical simulations.

Neuromuscular adaptations after a rehabilitation program in patients with chronic low back pain: case series (uncontrolled longitudinal study)

PubMed Central

2013-01-01

Background To investigate the impact of a short-term multimodal rehabilitation program for patients with low back pain (LBP) on trunk muscle reflex responses and feedforward activation induced by postural perturbations. Methods Case series (uncontrolled longitudinal study). Thirty chronic patients with LBP (21 women and 19 men, mean age 42.6 ± 8.6 years, mean weight 73 ± 14 kg, mean height 174 ± 10 cm) were included. The intervention consisted in a 5-day program including therapeutic education sessions (360 min), supervised abdominal and back muscle strength exercises (240 min), general aerobic training (150 min), stretching (150 min), postural education (150 min) and aqua therapy (150 min). Feedforward activation level and reflex amplitude determined by surface electromyographic activity triggered by postural perturbations were recorded from abdominal and paraspinal muscles in unexpected and expected conditions. Subjects were tested before, just after and again one month after the rehabilitation program. Results No main intervention effect was found on feedforward activation levels and reflex amplitudes underlining the absence of changes in the way patients with LBP reacted across perturbation conditions. However, we observed a shift in the behavioral strategy between conditions, in fact feedforward activation (similar in both conditions before the program) decreased in the unexpected condition after the program, whereas reflex amplitudes became similar in both conditions. Conclusions The results suggest that a short-term rehabilitation program modifies trunk behavioral strategies during postural perturbations. These results can be useful to clinicians for explaining to patients how to adapt to daily life activities before and after rehabilitation. PMID:24063646

Extracting functionally feedforward networks from a population of spiking neurons

PubMed Central

Vincent, Kathleen; Tauskela, Joseph S.; Thivierge, Jean-Philippe

2012-01-01

Neuronal avalanches are a ubiquitous form of activity characterized by spontaneous bursts whose size distribution follows a power-law. Recent theoretical models have replicated power-law avalanches by assuming the presence of functionally feedforward connections (FFCs) in the underlying dynamics of the system. Accordingly, avalanches are generated by a feedforward chain of activation that persists despite being embedded in a larger, massively recurrent circuit. However, it is unclear to what extent networks of living neurons that exhibit power-law avalanches rely on FFCs. Here, we employed a computational approach to reconstruct the functional connectivity of cultured cortical neurons plated on multielectrode arrays (MEAs) and investigated whether pharmacologically induced alterations in avalanche dynamics are accompanied by changes in FFCs. This approach begins by extracting a functional network of directed links between pairs of neurons, and then evaluates the strength of FFCs using Schur decomposition. In a first step, we examined the ability of this approach to extract FFCs from simulated spiking neurons. The strength of FFCs obtained in strictly feedforward networks diminished monotonically as links were gradually rewired at random. Next, we estimated the FFCs of spontaneously active cortical neuron cultures in the presence of either a control medium, a GABAA receptor antagonist (PTX), or an AMPA receptor antagonist combined with an NMDA receptor antagonist (APV/DNQX). The distribution of avalanche sizes in these cultures was modulated by this pharmacology, with a shallower power-law under PTX (due to the prominence of larger avalanches) and a steeper power-law under APV/DNQX (due to avalanches recruiting fewer neurons) relative to control cultures. The strength of FFCs increased in networks after application of PTX, consistent with an amplification of feedforward activity during avalanches. Conversely, FFCs decreased after application of APV/DNQX, consistent with fading feedforward activation. The observed alterations in FFCs provide experimental support for recent theoretical work linking power-law avalanches to the feedforward organization of functional connections in local neuronal circuits. PMID:23091458

Extracting functionally feedforward networks from a population of spiking neurons.

PubMed

Vincent, Kathleen; Tauskela, Joseph S; Thivierge, Jean-Philippe

2012-01-01

Neuronal avalanches are a ubiquitous form of activity characterized by spontaneous bursts whose size distribution follows a power-law. Recent theoretical models have replicated power-law avalanches by assuming the presence of functionally feedforward connections (FFCs) in the underlying dynamics of the system. Accordingly, avalanches are generated by a feedforward chain of activation that persists despite being embedded in a larger, massively recurrent circuit. However, it is unclear to what extent networks of living neurons that exhibit power-law avalanches rely on FFCs. Here, we employed a computational approach to reconstruct the functional connectivity of cultured cortical neurons plated on multielectrode arrays (MEAs) and investigated whether pharmacologically induced alterations in avalanche dynamics are accompanied by changes in FFCs. This approach begins by extracting a functional network of directed links between pairs of neurons, and then evaluates the strength of FFCs using Schur decomposition. In a first step, we examined the ability of this approach to extract FFCs from simulated spiking neurons. The strength of FFCs obtained in strictly feedforward networks diminished monotonically as links were gradually rewired at random. Next, we estimated the FFCs of spontaneously active cortical neuron cultures in the presence of either a control medium, a GABA(A) receptor antagonist (PTX), or an AMPA receptor antagonist combined with an NMDA receptor antagonist (APV/DNQX). The distribution of avalanche sizes in these cultures was modulated by this pharmacology, with a shallower power-law under PTX (due to the prominence of larger avalanches) and a steeper power-law under APV/DNQX (due to avalanches recruiting fewer neurons) relative to control cultures. The strength of FFCs increased in networks after application of PTX, consistent with an amplification of feedforward activity during avalanches. Conversely, FFCs decreased after application of APV/DNQX, consistent with fading feedforward activation. The observed alterations in FFCs provide experimental support for recent theoretical work linking power-law avalanches to the feedforward organization of functional connections in local neuronal circuits.

Combined line-of-sight error and angular position to generate feedforward control for a charge-coupled device-based tracking loop

NASA Astrophysics Data System (ADS)

Tang, Tao; Cai, Huaxiang; Huang, Yongmei; Ren, Ge

2015-10-01

A feedforward control based on data fusion is proposed to enhance closed-loop performance. The target trajectory as the observed value of a Kalman filter is recovered by synthesizing line-of-sight error and angular position from the encoder. A Kalman filter based on a Singer acceleration model is employed to estimate the target velocity. In this control scheme, the control stability is influenced by the bandwidth of the Kalman filter and time misalignment. The transfer function of the Kalman filter in the frequency domain is built for analyzing the closed loop stability, which shows that the Kalman filter is the major factor that affects the control stability. The feedforward control proposed here is verified through simulations and experiments.

Improving atomic force microscopy imaging by a direct inverse asymmetric PI hysteresis model.

PubMed

Wang, Dong; Yu, Peng; Wang, Feifei; Chan, Ho-Yin; Zhou, Lei; Dong, Zaili; Liu, Lianqing; Li, Wen Jung

2015-02-03

A modified Prandtl-Ishlinskii (PI) model, referred to as a direct inverse asymmetric PI (DIAPI) model in this paper, was implemented to reduce the displacement error between a predicted model and the actual trajectory of a piezoelectric actuator which is commonly found in AFM systems. Due to the nonlinearity of the piezoelectric actuator, the standard symmetric PI model cannot precisely describe the asymmetric motion of the actuator. In order to improve the accuracy of AFM scans, two series of slope parameters were introduced in the PI model to describe both the voltage-increase-loop (trace) and voltage-decrease-loop (retrace). A feedforward controller based on the DIAPI model was implemented to compensate hysteresis. Performance of the DIAPI model and the feedforward controller were validated by scanning micro-lenses and standard silicon grating using a custom-built AFM.

Feedforward and feedback frequency-dependent interactions in a large-scale laminar network of the primate cortex.

PubMed

Mejias, Jorge F; Murray, John D; Kennedy, Henry; Wang, Xiao-Jing

2016-11-01

Interactions between top-down and bottom-up processes in the cerebral cortex hold the key to understanding attentional processes, predictive coding, executive control, and a gamut of other brain functions. However, the underlying circuit mechanism remains poorly understood and represents a major challenge in neuroscience. We approached this problem using a large-scale computational model of the primate cortex constrained by new directed and weighted connectivity data. In our model, the interplay between feedforward and feedback signaling depends on the cortical laminar structure and involves complex dynamics across multiple (intralaminar, interlaminar, interareal, and whole cortex) scales. The model was tested by reproducing, as well as providing insights into, a wide range of neurophysiological findings about frequency-dependent interactions between visual cortical areas, including the observation that feedforward pathways are associated with enhanced gamma (30 to 70 Hz) oscillations, whereas feedback projections selectively modulate alpha/low-beta (8 to 15 Hz) oscillations. Furthermore, the model reproduces a functional hierarchy based on frequency-dependent Granger causality analysis of interareal signaling, as reported in recent monkey and human experiments, and suggests a mechanism for the observed context-dependent hierarchy dynamics. Together, this work highlights the necessity of multiscale approaches and provides a modeling platform for studies of large-scale brain circuit dynamics and functions.

Feedforward and feedback frequency-dependent interactions in a large-scale laminar network of the primate cortex

PubMed Central

Mejias, Jorge F.; Murray, John D.; Kennedy, Henry; Wang, Xiao-Jing

2016-01-01

Interactions between top-down and bottom-up processes in the cerebral cortex hold the key to understanding attentional processes, predictive coding, executive control, and a gamut of other brain functions. However, the underlying circuit mechanism remains poorly understood and represents a major challenge in neuroscience. We approached this problem using a large-scale computational model of the primate cortex constrained by new directed and weighted connectivity data. In our model, the interplay between feedforward and feedback signaling depends on the cortical laminar structure and involves complex dynamics across multiple (intralaminar, interlaminar, interareal, and whole cortex) scales. The model was tested by reproducing, as well as providing insights into, a wide range of neurophysiological findings about frequency-dependent interactions between visual cortical areas, including the observation that feedforward pathways are associated with enhanced gamma (30 to 70 Hz) oscillations, whereas feedback projections selectively modulate alpha/low-beta (8 to 15 Hz) oscillations. Furthermore, the model reproduces a functional hierarchy based on frequency-dependent Granger causality analysis of interareal signaling, as reported in recent monkey and human experiments, and suggests a mechanism for the observed context-dependent hierarchy dynamics. Together, this work highlights the necessity of multiscale approaches and provides a modeling platform for studies of large-scale brain circuit dynamics and functions. PMID:28138530

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

Childs, Andrew M.; Center for Theoretical Physics, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139; Leung, Debbie W.

We present unified, systematic derivations of schemes in the two known measurement-based models of quantum computation. The first model (introduced by Raussendorf and Briegel, [Phys. Rev. Lett. 86, 5188 (2001)]) uses a fixed entangled state, adaptive measurements on single qubits, and feedforward of the measurement results. The second model (proposed by Nielsen, [Phys. Lett. A 308, 96 (2003)] and further simplified by Leung, [Int. J. Quant. Inf. 2, 33 (2004)]) uses adaptive two-qubit measurements that can be applied to arbitrary pairs of qubits, and feedforward of the measurement results. The underlying principle of our derivations is a variant of teleportationmore » introduced by Zhou, Leung, and Chuang, [Phys. Rev. A 62, 052316 (2000)]. Our derivations unify these two measurement-based models of quantum computation and provide significantly simpler schemes.« less

Reliability analysis of C-130 turboprop engine components using artificial neural network

NASA Astrophysics Data System (ADS)

Qattan, Nizar A.

In this study, we predict the failure rate of Lockheed C-130 Engine Turbine. More than thirty years of local operational field data were used for failure rate prediction and validation. The Weibull regression model and the Artificial Neural Network model including (feed-forward back-propagation, radial basis neural network, and multilayer perceptron neural network model); will be utilized to perform this study. For this purpose, the thesis will be divided into five major parts. First part deals with Weibull regression model to predict the turbine general failure rate, and the rate of failures that require overhaul maintenance. The second part will cover the Artificial Neural Network (ANN) model utilizing the feed-forward back-propagation algorithm as a learning rule. The MATLAB package will be used in order to build and design a code to simulate the given data, the inputs to the neural network are the independent variables, the output is the general failure rate of the turbine, and the failures which required overhaul maintenance. In the third part we predict the general failure rate of the turbine and the failures which require overhaul maintenance, using radial basis neural network model on MATLAB tool box. In the fourth part we compare the predictions of the feed-forward back-propagation model, with that of Weibull regression model, and radial basis neural network model. The results show that the failure rate predicted by the feed-forward back-propagation artificial neural network model is closer in agreement with radial basis neural network model compared with the actual field-data, than the failure rate predicted by the Weibull model. By the end of the study, we forecast the general failure rate of the Lockheed C-130 Engine Turbine, the failures which required overhaul maintenance and six categorical failures using multilayer perceptron neural network (MLP) model on DTREG commercial software. The results also give an insight into the reliability of the engine turbine under actual operating conditions, which can be used by aircraft operators for assessing system and component failures and customizing the maintenance programs recommended by the manufacturer.

Artificial neural network modelling of uncertainty in gamma-ray spectrometry

NASA Astrophysics Data System (ADS)

Dragović, S.; Onjia, A.; Stanković, S.; Aničin, I.; Bačić, G.

2005-03-01

An artificial neural network (ANN) model for the prediction of measuring uncertainties in gamma-ray spectrometry was developed and optimized. A three-layer feed-forward ANN with back-propagation learning algorithm was used to model uncertainties of measurement of activity levels of eight radionuclides ( 226Ra, 238U, 235U, 40K, 232Th, 134Cs, 137Cs and 7Be) in soil samples as a function of measurement time. It was shown that the neural network provides useful data even from small experimental databases. The performance of the optimized neural network was found to be very good, with correlation coefficients ( R2) between measured and predicted uncertainties ranging from 0.9050 to 0.9915. The correlation coefficients did not significantly deteriorate when the network was tested on samples with greatly different uranium-to-thorium ( 238U/ 232Th) ratios. The differences between measured and predicted uncertainties were not influenced by the absolute values of uncertainties of measured radionuclide activities. Once the ANN is trained, it could be employed in analyzing soil samples regardless of the 238U/ 232Th ratio. It was concluded that a considerable saving in time could be obtained using the trained neural network model for predicting the measurement times needed to attain the desired statistical accuracy.

Prediction of phycoremediation of As(III) and As(V) from synthetic wastewater by Chlorella pyrenoidosa using artificial neural network

NASA Astrophysics Data System (ADS)

Podder, M. S.; Majumder, C. B.

2017-11-01

An artificial neural network (ANN) model was developed to predict the phycoremediation efficiency of Chlorella pyrenoidosa for the removal of both As(III) and As(V) from synthetic wastewater based on 49 data-sets obtained from experimental study and increased the data using CSCF technique. The data were divided into training (60%) validation (20%) and testing (20%) sets. The data collected was used for training a three-layer feed-forward back propagation (BP) learning algorithm having 4-5-1 architecture. The model used tangent sigmoid transfer function at input to hidden layer ( tansing) while a linear transfer function ( purelin) was used at output layer. Comparison between experimental results and model results gave a high correlation coefficient (R allANN 2 equal to 0.99987 for both ions and exhibited that the model was able to predict the phycoremediation of As(III) and As(V) from wastewater. Experimental parameters influencing phycoremediation process like pH, inoculum size, contact time and initial arsenic concentration [either As(III) or As(V)] were investigated. A contact time of 168 h was mainly required for achieving equilibrium at pH 9.0 with an inoculum size of 10% (v/v). At optimum conditions, metal ion uptake enhanced with increasing initial metal ion concentration.

Smooth function approximation using neural networks.

PubMed

Ferrari, Silvia; Stengel, Robert F

2005-01-01

An algebraic approach for representing multidimensional nonlinear functions by feedforward neural networks is presented. In this paper, the approach is implemented for the approximation of smooth batch data containing the function's input, output, and possibly, gradient information. The training set is associated to the network adjustable parameters by nonlinear weight equations. The cascade structure of these equations reveals that they can be treated as sets of linear systems. Hence, the training process and the network approximation properties can be investigated via linear algebra. Four algorithms are developed to achieve exact or approximate matching of input-output and/or gradient-based training sets. Their application to the design of forward and feedback neurocontrollers shows that algebraic training is characterized by faster execution speeds and better generalization properties than contemporary optimization techniques.

Hardware-Efficient On-line Learning through Pipelined Truncated-Error Backpropagation in Binary-State Networks

PubMed Central

Mostafa, Hesham; Pedroni, Bruno; Sheik, Sadique; Cauwenberghs, Gert

2017-01-01

Artificial neural networks (ANNs) trained using backpropagation are powerful learning architectures that have achieved state-of-the-art performance in various benchmarks. Significant effort has been devoted to developing custom silicon devices to accelerate inference in ANNs. Accelerating the training phase, however, has attracted relatively little attention. In this paper, we describe a hardware-efficient on-line learning technique for feedforward multi-layer ANNs that is based on pipelined backpropagation. Learning is performed in parallel with inference in the forward pass, removing the need for an explicit backward pass and requiring no extra weight lookup. By using binary state variables in the feedforward network and ternary errors in truncated-error backpropagation, the need for any multiplications in the forward and backward passes is removed, and memory requirements for the pipelining are drastically reduced. Further reduction in addition operations owing to the sparsity in the forward neural and backpropagating error signal paths contributes to highly efficient hardware implementation. For proof-of-concept validation, we demonstrate on-line learning of MNIST handwritten digit classification on a Spartan 6 FPGA interfacing with an external 1Gb DDR2 DRAM, that shows small degradation in test error performance compared to an equivalently sized binary ANN trained off-line using standard back-propagation and exact errors. Our results highlight an attractive synergy between pipelined backpropagation and binary-state networks in substantially reducing computation and memory requirements, making pipelined on-line learning practical in deep networks. PMID:28932180

Hardware-Efficient On-line Learning through Pipelined Truncated-Error Backpropagation in Binary-State Networks.

PubMed

Mostafa, Hesham; Pedroni, Bruno; Sheik, Sadique; Cauwenberghs, Gert

2017-01-01

Artificial neural networks (ANNs) trained using backpropagation are powerful learning architectures that have achieved state-of-the-art performance in various benchmarks. Significant effort has been devoted to developing custom silicon devices to accelerate inference in ANNs. Accelerating the training phase, however, has attracted relatively little attention. In this paper, we describe a hardware-efficient on-line learning technique for feedforward multi-layer ANNs that is based on pipelined backpropagation. Learning is performed in parallel with inference in the forward pass, removing the need for an explicit backward pass and requiring no extra weight lookup. By using binary state variables in the feedforward network and ternary errors in truncated-error backpropagation, the need for any multiplications in the forward and backward passes is removed, and memory requirements for the pipelining are drastically reduced. Further reduction in addition operations owing to the sparsity in the forward neural and backpropagating error signal paths contributes to highly efficient hardware implementation. For proof-of-concept validation, we demonstrate on-line learning of MNIST handwritten digit classification on a Spartan 6 FPGA interfacing with an external 1Gb DDR2 DRAM, that shows small degradation in test error performance compared to an equivalently sized binary ANN trained off-line using standard back-propagation and exact errors. Our results highlight an attractive synergy between pipelined backpropagation and binary-state networks in substantially reducing computation and memory requirements, making pipelined on-line learning practical in deep networks.

Computational mechanisms underlying cortical responses to the affordance properties of visual scenes

PubMed Central

Epstein, Russell A.

2018-01-01

Biologically inspired deep convolutional neural networks (CNNs), trained for computer vision tasks, have been found to predict cortical responses with remarkable accuracy. However, the internal operations of these models remain poorly understood, and the factors that account for their success are unknown. Here we develop a set of techniques for using CNNs to gain insights into the computational mechanisms underlying cortical responses. We focused on responses in the occipital place area (OPA), a scene-selective region of dorsal occipitoparietal cortex. In a previous study, we showed that fMRI activation patterns in the OPA contain information about the navigational affordances of scenes; that is, information about where one can and cannot move within the immediate environment. We hypothesized that this affordance information could be extracted using a set of purely feedforward computations. To test this idea, we examined a deep CNN with a feedforward architecture that had been previously trained for scene classification. We found that responses in the CNN to scene images were highly predictive of fMRI responses in the OPA. Moreover the CNN accounted for the portion of OPA variance relating to the navigational affordances of scenes. The CNN could thus serve as an image-computable candidate model of affordance-related responses in the OPA. We then ran a series of in silico experiments on this model to gain insights into its internal operations. These analyses showed that the computation of affordance-related features relied heavily on visual information at high-spatial frequencies and cardinal orientations, both of which have previously been identified as low-level stimulus preferences of scene-selective visual cortex. These computations also exhibited a strong preference for information in the lower visual field, which is consistent with known retinotopic biases in the OPA. Visualizations of feature selectivity within the CNN suggested that affordance-based responses encoded features that define the layout of the spatial environment, such as boundary-defining junctions and large extended surfaces. Together, these results map the sensory functions of the OPA onto a fully quantitative model that provides insights into its visual computations. More broadly, they advance integrative techniques for understanding visual cortex across multiple level of analysis: from the identification of cortical sensory functions to the modeling of their underlying algorithms. PMID:29684011

Hysteresis compensation for piezoelectric actuators in single-point diamond turning

NASA Astrophysics Data System (ADS)

Wang, Haifeng; Hu, Dejin; Wan, Daping; Liu, Hongbin

2006-02-01

In recent years, interests have been growing for fast tool servo (FTS) systems to increase the capability of existing single-point diamond turning machines. Although piezoelectric actuator is the most universal base of FTS system due to its high stiffness, accuracy and bandwidth, nonlinearity in piezoceramics limits both the static and dynamic performance of piezoelectric-actuated control systems evidently. To compensate the nonlinear hysteresis behavior of piezoelectric actuators, a hybrid model coupled with Preisach model and feedforward neural network (FNN) has been described. Since the training of FNN does not require a special calibration sequence, it is possible for on-line identification and real-time implementation with general operating data of a specific piezoelectric actuator. To describe the rate dependent behavior of piezoelectric actuators, a hybrid dynamic model was developed to predict the response of piezoelectric actuators in a wider range of input frequency. Experimental results show that a maximal error of less than 3% was accomplished by this dynamic model.

Application of a passivity based control methodology for flexible joint robots to a simplified Space Shuttle RMS

NASA Technical Reports Server (NTRS)

Sicard, Pierre; Wen, John T.

1992-01-01

A passivity approach for the control design of flexible joint robots is applied to the rate control of a three-link arm modeled after the shoulder yaw joint of the Space Shuttle Remote Manipulator System (RMS). The system model includes friction and elastic joint couplings modeled as nonlinear springs. The basic structure of the proposed controller is the sum of a model-based feedforward and a model-independent feedback. A regulator approach with link state feedback is employed to define the desired motor state. Passivity theory is used to design a motor state-based controller to stabilize the error system formed by the feedforward. Simulation results show that greatly improved performance was obtained by using the proposed controller over the existing RMS controller.

Phase I to II cross-induction of xenobiotic metabolizing enzymes: a feedforward control mechanism for potential hormetic responses.

PubMed

Zhang, Qiang; Pi, Jingbo; Woods, Courtney G; Andersen, Melvin E

2009-06-15

Hormetic responses to xenobiotic exposure likely occur as a result of overcompensation by the homeostatic control systems operating in biological organisms. However, the mechanisms underlying overcompensation that leads to hormesis are still unclear. A well-known homeostatic circuit in the cell is the gene induction network comprising phase I, II and III metabolizing enzymes, which are responsible for xenobiotic detoxification, and in many cases, bioactivation. By formulating a differential equation-based computational model, we investigated in this study whether hormesis can arise from the operation of this gene/enzyme network. The model consists of two feedback and one feedforward controls. With the phase I negative feedback control, xenobiotic X activates nuclear receptors to induce cytochrome P450 enzyme, which bioactivates X into a reactive metabolite X'. With the phase II negative feedback control, X' activates transcription factor Nrf2 to induce phase II enzymes such as glutathione S-transferase and glutamate cysteine ligase, etc., which participate in a set of reactions that lead to the metabolism of X' into a less toxic conjugate X''. The feedforward control involves phase I to II cross-induction, in which the parent chemical X can also induce phase II enzymes directly through the nuclear receptor and indirectly through transcriptionally upregulating Nrf2. As a result of the active feedforward control, a steady-state hormetic relationship readily arises between the concentrations of the reactive metabolite X' and the extracellular parent chemical X to which the cell is exposed. The shape of dose-response evolves over time from initially monotonically increasing to J-shaped at the final steady state-a temporal sequence consistent with adaptation-mediated hormesis. The magnitude of the hormetic response is enhanced by increases in the feedforward gain, but attenuated by increases in the bioactivation or phase II feedback loop gains. Our study suggests a possibly common mechanism for the hormetic responses observed with many mutagens/carcinogens whose activities require bioactivation by phase I enzymes. Feedforward control, often operating in combination with negative feedback regulation in a homeostatic system, may be a general control theme responsible for steady-state hormesis.

Startle reduces recall of a recently learned internal model.

PubMed

Wright, Zachary; Patton, James L; Ravichandran, Venn

2011-01-01

Recent work has shown that preplanned motor programs are released early from subcortical areas by the using a startling acoustic stimulus (SAS). Our question is whether this response might also contain a recently learned internal model, which draws on experience to predict and compensate for expected perturbations in a feedforward manner. Studies of adaptation to robotic forces have shown some evidence of this, but were potentially confounded by cocontraction caused by startle. We performed a new adaptation experiment using a visually distorted field that could not be confounded by cocontraction. We found that in all subjects that exhibited startle, the startle stimulus (1) reduced performance of the recently learned task (2) reduced after-effect magnitudes. Because startle reduced but did not eliminate the recall of learned control, we suggest that multiple neural centers (cortical and subcortical) are involved in such learning and adaptation, which can impact training areas such as piloting, teleoperation, sports, and rehabilitation. © 2011 IEEE

Monthly monsoon rainfall forecasting using artificial neural networks

NASA Astrophysics Data System (ADS)

Ganti, Ravikumar

2014-10-01

Indian agriculture sector heavily depends on monsoon rainfall for successful harvesting. In the past, prediction of rainfall was mainly performed using regression models, which provide reasonable accuracy in the modelling and forecasting of complex physical systems. Recently, Artificial Neural Networks (ANNs) have been proposed as efficient tools for modelling and forecasting. A feed-forward multi-layer perceptron type of ANN architecture trained using the popular back-propagation algorithm was employed in this study. Other techniques investigated for modeling monthly monsoon rainfall include linear and non-linear regression models for comparison purposes. The data employed in this study include monthly rainfall and monthly average of the daily maximum temperature in the North Central region in India. Specifically, four regression models and two ANN model's were developed. The performance of various models was evaluated using a wide variety of standard statistical parameters and scatter plots. The results obtained in this study for forecasting monsoon rainfalls using ANNs have been encouraging. India's economy and agricultural activities can be effectively managed with the help of the availability of the accurate monsoon rainfall forecasts.

Intelligent processing of acoustic emission signals

NASA Astrophysics Data System (ADS)

Sachse, Wolfgang; Grabec, Igor

1992-07-01

Recent developments in applying neural-like signal-processing procedures for analyzing acoustic emission signals are summarized. These procedures employ a set of learning signals to develop a memory that can subsequently be utilized to process other signals to recover information about an unknown source. A majority of the current applications to process ultrasonic waveforms are based on multilayered, feed-forward neural networks, trained with some type of back-error propagation rule.

Using statistical and artificial neural network models to forecast potentiometric levels at a deep well in South Texas

NASA Astrophysics Data System (ADS)

Uddameri, V.

2007-01-01

Reliable forecasts of monthly and quarterly fluctuations in groundwater levels are necessary for short- and medium-term planning and management of aquifers to ensure proper service of seasonal demands within a region. Development of physically based transient mathematical models at this time scale poses considerable challenges due to lack of suitable data and other uncertainties. Artificial neural networks (ANN) possess flexible mathematical structures and are capable of mapping highly nonlinear relationships. Feed-forward neural network models were constructed and trained using the back-percolation algorithm to forecast monthly and quarterly time-series water levels at a well that taps into the deeper Evangeline formation of the Gulf Coast aquifer in Victoria, TX. Unlike unconfined formations, no causal relationships exist between water levels and hydro-meteorological variables measured near the vicinity of the well. As such, an endogenous forecasting model using dummy variables to capture short-term seasonal fluctuations and longer-term (decadal) trends was constructed. The root mean square error, mean absolute deviation and correlation coefficient ( R) were noted to be 1.40, 0.33 and 0.77 m, respectively, for an evaluation dataset of quarterly measurements and 1.17, 0.46, and 0.88 m for an evaluative monthly dataset not used to train or test the model. These statistics were better for the ANN model than those developed using statistical regression techniques.

Coordinated three-dimensional motion of the head and torso by dynamic neural networks.

PubMed

Kim, J; Hemami, H

1998-01-01

The problem of trajectory tracking control of a three dimensional (3D) model of the human upper torso and head is considered. The torso and the head are modeled as two rigid bodies connected at one point, and the Newton-Euler method is used to derive the nonlinear differential equations that govern the motion of the system. The two-link system is driven by six pairs of muscle like actuators that possess physiologically inspired alpha like and gamma like inputs, and spindle like and Golgi tendon organ like outputs. These outputs are utilized as reflex feedback for stability and stiffness control, in a long loop feedback for the purpose of estimating the state of the system (somesthesis), and as part of the input to the controller. Ideal delays of different duration are included in the feedforward and feedback paths of the system to emulate such delays encountered in physiological systems. Dynamical neural networks are trained to learn effective control of the desired maneuvers of the system. The feasibility of the controller is demonstrated by computer simulation of the successful execution of the desired maneuvers. This work demonstrates the capabilities of neural circuits in controlling highly nonlinear systems with multidelays in their feedforward and feedback paths. The ultimate long range goal of this research is toward understanding the working of the central nervous system in controlling movement. It is an interdisciplinary effort relying on mechanics, biomechanics, neuroscience, system theory, physiology and anatomy, and its short range relevance to rehabilitation must be noted.

The neural representation of the gender of faces in the primate visual system: A computer modeling study.

PubMed

Minot, Thomas; Dury, Hannah L; Eguchi, Akihiro; Humphreys, Glyn W; Stringer, Simon M

2017-03-01

We use an established neural network model of the primate visual system to show how neurons might learn to encode the gender of faces. The model consists of a hierarchy of 4 competitive neuronal layers with associatively modifiable feedforward synaptic connections between successive layers. During training, the network was presented with many realistic images of male and female faces, during which the synaptic connections are modified using biologically plausible local associative learning rules. After training, we found that different subsets of output neurons have learned to respond exclusively to either male or female faces. With the inclusion of short range excitation within each neuronal layer to implement a self-organizing map architecture, neurons representing either male or female faces were clustered together in the output layer. This learning process is entirely unsupervised, as the gender of the face images is not explicitly labeled and provided to the network as a supervisory training signal. These simulations are extended to training the network on rotating faces. It is found that by using a trace learning rule incorporating a temporal memory trace of recent neuronal activity, neurons responding selectively to either male or female faces were also able to learn to respond invariantly over different views of the faces. This kind of trace learning has been previously shown to operate within the primate visual system by neurophysiological and psychophysical studies. The computer simulations described here predict that similar neurons encoding the gender of faces will be present within the primate visual system. (PsycINFO Database Record (c) 2017 APA, all rights reserved).

Feedforward operation of a lens setup for large defocus and astigmatism correction

NASA Astrophysics Data System (ADS)

Verstraete, Hans R. G. W.; Almasian, MItra; Pozzi, Paolo; Bilderbeek, Rolf; Kalkman, Jeroen; Faber, Dirk J.; Verhaegen, Michel

2016-04-01

In this manuscript, we present a lens setup for large defocus and astigmatism correction. A deformable defocus lens and two rotational cylindrical lenses are used to control the defocus and astigmatism. The setup is calibrated using a simple model that allows the calculation of the lens inputs so that a desired defocus and astigmatism are actuated on the eye. The setup is tested by determining the feedforward prediction error, imaging a resolution target, and removing introduced aberrations.

Stochastic Feedforward Control Technique

NASA Technical Reports Server (NTRS)

Halyo, Nesim

1990-01-01

Class of commanded trajectories modeled as stochastic process. Advanced Transport Operating Systems (ATOPS) research and development program conducted by NASA Langley Research Center aimed at developing capabilities for increases in capacities of airports, safe and accurate flight in adverse weather conditions including shear, winds, avoidance of wake vortexes, and reduced consumption of fuel. Advances in techniques for design of modern controls and increased capabilities of digital flight computers coupled with accurate guidance information from Microwave Landing System (MLS). Stochastic feedforward control technique developed within context of ATOPS program.

High-speed tracking control of piezoelectric actuators using an ellipse-based hysteresis model.

PubMed

Gu, Guoying; Zhu, Limin

2010-08-01

In this paper, an ellipse-based mathematic model is developed to characterize the rate-dependent hysteresis in piezoelectric actuators. Based on the proposed model, an expanded input space is constructed to describe the multivalued hysteresis function H[u](t) by a multiple input single output (MISO) mapping Gamma:R(2)-->R. Subsequently, the inverse MISO mapping Gamma(-1)(H[u](t),H[u](t);u(t)) is proposed for real-time hysteresis compensation. In controller design, a hybrid control strategy combining a model-based feedforward controller and a proportional integral differential (PID) feedback loop is used for high-accuracy and high-speed tracking control of piezoelectric actuators. The real-time feedforward controller is developed to cancel the rate-dependent hysteresis based on the inverse hysteresis model, while the PID controller is used to compensate for the creep, modeling errors, and parameter uncertainties. Finally, experiments with and without hysteresis compensation are conducted and the experimental results are compared. The experimental results show that the hysteresis compensation in the feedforward path can reduce the hysteresis-caused error by up to 88% and the tracking performance of the hybrid controller is greatly improved in high-speed tracking control applications, e.g., the root-mean-square tracking error is reduced to only 0.34% of the displacement range under the input frequency of 100 Hz.

Output control using feedforward and cascade controllers

NASA Technical Reports Server (NTRS)

Seraji, H.

1987-01-01

An open-loop solution to the output control problem in SISO (single-input, single-output) systems by means of feedforward and cascade controllers is investigated. A simple characterization of feedforward controllers, which achieve steady-state disturbance rejection, is given in a transfer-function setting. Cascade controllers which cause steady-state command tracking are characterized. Disturbance decoupling and command matching controllers are identified. Conditions for existence of feedforward and cascade controllers are given. For unstable systems, it is shown that a stabilizing feedback controller can be used without affecting the feedforward and cascade controllers used for output control; hence, the three controllers can be designed independently. Output control by a combination of feedforward and feedback is discussed.

Method and apparatus for adaptive force and position control of manipulators

NASA Technical Reports Server (NTRS)

Seraji, Homayoun (Inventor)

1989-01-01

The present invention discloses systematic methods and apparatus for the design of real time controllers. Real-time control employs adaptive force/position by use of feedforward and feedback controllers, with the feedforward controller being the inverse of the linearized model of robot dynamics and containing only proportional-double-derivative terms is disclosed. The feedback controller, of the proportional-integral-derivative type, ensures that manipulator joints follow reference trajectories and the feedback controller achieves robust tracking of step-plus-exponential trajectories, all in real time. The adaptive controller includes adaptive force and position control within a hybrid control architecture. The adaptive controller, for force control, achieves tracking of desired force setpoints, and the adaptive position controller accomplishes tracking of desired position trajectories. Circuits in the adaptive feedback and feedforward controllers are varied by adaptation laws.

Development of feedforward control in a dynamic manual tracking task.

PubMed

van Roon, Dominique; Caeyenberghs, Karen; Swinnen, Stephan P; Smits-Engelsman, Bouwien C M

2008-01-01

To examine the development of feedforward control during manual tracking, 117 participants in 5 age groups (6 to 7, 8 to 9, 10 to 11, 12 to 14, and 15 to 17 years) tracked an accelerating dot presented on a monitor by moving an electronic pen on a digitizer. To remain successful at higher target velocities, they had to create a predictive model of the target's motion. The ability to track the target at higher velocities increased, and the application of a feedback-based step-and-hold strategy decreased with age, as shown by increases in maximum target velocity and decreases in number of stops between ages 6-7 and 8-9 and between ages 8-9 and 10-11. The ability to exploit feedforward control in a dynamic tracking task improves significantly with age.

Neural network for image compression

NASA Astrophysics Data System (ADS)

Panchanathan, Sethuraman; Yeap, Tet H.; Pilache, B.

1992-09-01

In this paper, we propose a new scheme for image compression using neural networks. Image data compression deals with minimization of the amount of data required to represent an image while maintaining an acceptable quality. Several image compression techniques have been developed in recent years. We note that the coding performance of these techniques may be improved by employing adaptivity. Over the last few years neural network has emerged as an effective tool for solving a wide range of problems involving adaptivity and learning. A multilayer feed-forward neural network trained using the backward error propagation algorithm is used in many applications. However, this model is not suitable for image compression because of its poor coding performance. Recently, a self-organizing feature map (SOFM) algorithm has been proposed which yields a good coding performance. However, this algorithm requires a long training time because the network starts with random initial weights. In this paper we have used the backward error propagation algorithm (BEP) to quickly obtain the initial weights which are then used to speedup the training time required by the SOFM algorithm. The proposed approach (BEP-SOFM) combines the advantages of the two techniques and, hence, achieves a good coding performance in a shorter training time. Our simulation results demonstrate the potential gains using the proposed technique.

Detection of different states of sleep in the rodents by the means of artificial neural networks

NASA Astrophysics Data System (ADS)

Musatov, Viacheslav; Dykin, Viacheslav; Pitsik, Elena; Pisarchik, Alexander

2018-04-01

This paper considers the possibility of classification of electroencephalogram (EEG) and electromyogram (EMG) signals corresponding to different phases of sleep and wakefulness of mice by the means of artificial neural networks. A feed-forward artificial neural network based on multilayer perceptron was created and trained on the data of one of the rodents. The trained network was used to read and classify the EEG and EMG data corresponding to different phases of sleep and wakefulness of the same mouse and other mouse. The results show a good recognition quality of all phases for the rodent on which the training was conducted (80-99%) and acceptable recognition quality for the data collected from the same mouse after a stroke.

A Robust Feedforward Model of the Olfactory System

NASA Astrophysics Data System (ADS)

Zhang, Yilun; Sharpee, Tatyana

Most natural odors have sparse molecular composition. This makes the principles of compressing sensing potentially relevant to the structure of the olfactory code. Yet, the largely feedforward organization of the olfactory system precludes reconstruction using standard compressed sensing algorithms. To resolve this problem, recent theoretical work has proposed that signal reconstruction could take place as a result of a low dimensional dynamical system converging to one of its attractor states. The dynamical aspects of optimization, however, would slow down odor recognition and were also found to be susceptible to noise. Here we describe a feedforward model of the olfactory system that achieves both strong compression and fast reconstruction that is also robust to noise. A key feature of the proposed model is a specific relationship between how odors are represented at the glomeruli stage, which corresponds to a compression, and the connections from glomeruli to Kenyon cells, which in the model corresponds to reconstruction. We show that provided this specific relationship holds true, the reconstruction will be both fast and robust to noise, and in particular to failure of glomeruli. The predicted connectivity rate from glomeruli to the Kenyon cells can be tested experimentally. This research was supported by James S. McDonnell Foundation, NSF CAREER award IIS-1254123, NSF Ideas Lab Collaborative Research IOS 1556388.

Improving Atomic Force Microscopy Imaging by a Direct Inverse Asymmetric PI Hysteresis Model

PubMed Central

Wang, Dong; Yu, Peng; Wang, Feifei; Chan, Ho-Yin; Zhou, Lei; Dong, Zaili; Liu, Lianqing; Li, Wen Jung

2015-01-01

A modified Prandtl–Ishlinskii (PI) model, referred to as a direct inverse asymmetric PI (DIAPI) model in this paper, was implemented to reduce the displacement error between a predicted model and the actual trajectory of a piezoelectric actuator which is commonly found in AFM systems. Due to the nonlinearity of the piezoelectric actuator, the standard symmetric PI model cannot precisely describe the asymmetric motion of the actuator. In order to improve the accuracy of AFM scans, two series of slope parameters were introduced in the PI model to describe both the voltage-increase-loop (trace) and voltage-decrease-loop (retrace). A feedforward controller based on the DIAPI model was implemented to compensate hysteresis. Performance of the DIAPI model and the feedforward controller were validated by scanning micro-lenses and standard silicon grating using a custom-built AFM. PMID:25654719

Parvalbumin interneuron mediated feedforward inhibition controls signal output in the deep layers of the perirhinal‐entorhinal cortex

PubMed Central

Willems, Janske G. P.; Wadman, Wytse J.

2018-01-01

Abstract The perirhinal (PER) and lateral entorhinal (LEC) cortex form an anatomical link between the neocortex and the hippocampus. However, neocortical activity is transmitted through the PER and LEC to the hippocampus with a low probability, suggesting the involvement of the inhibitory network. This study explored the role of interneuron mediated inhibition, activated by electrical stimulation in the agranular insular cortex (AiP), in the deep layers of the PER and LEC. Activated synaptic input by AiP stimulation rarely evoked action potentials in the PER‐LEC deep layer excitatory principal neurons, most probably because the evoked synaptic response consisted of a small excitatory and large inhibitory conductance. Furthermore, parvalbumin positive (PV) interneurons—a subset of interneurons projecting onto the axo‐somatic region of principal neurons—received synaptic input earlier than principal neurons, suggesting recruitment of feedforward inhibition. This synaptic input in PV interneurons evoked varying trains of action potentials, explaining the fast rising, long lasting synaptic inhibition received by deep layer principal neurons. Altogether, the excitatory input from the AiP onto deep layer principal neurons is overruled by strong feedforward inhibition. PV interneurons, with their fast, extensive stimulus‐evoked firing, are able to deliver this fast evoked inhibition in principal neurons. This indicates an essential role for PV interneurons in the gating mechanism of the PER‐LEC network. PMID:29341361

On adaptive learning rate that guarantees convergence in feedforward networks.

PubMed

Behera, Laxmidhar; Kumar, Swagat; Patnaik, Awhan

2006-09-01

This paper investigates new learning algorithms (LF I and LF II) based on Lyapunov function for the training of feedforward neural networks. It is observed that such algorithms have interesting parallel with the popular backpropagation (BP) algorithm where the fixed learning rate is replaced by an adaptive learning rate computed using convergence theorem based on Lyapunov stability theory. LF II, a modified version of LF I, has been introduced with an aim to avoid local minima. This modification also helps in improving the convergence speed in some cases. Conditions for achieving global minimum for these kind of algorithms have been studied in detail. The performances of the proposed algorithms are compared with BP algorithm and extended Kalman filtering (EKF) on three bench-mark function approximation problems: XOR, 3-bit parity, and 8-3 encoder. The comparisons are made in terms of number of learning iterations and computational time required for convergence. It is found that the proposed algorithms (LF I and II) are much faster in convergence than other two algorithms to attain same accuracy. Finally, the comparison is made on a complex two-dimensional (2-D) Gabor function and effect of adaptive learning rate for faster convergence is verified. In a nutshell, the investigations made in this paper help us better understand the learning procedure of feedforward neural networks in terms of adaptive learning rate, convergence speed, and local minima.

Rapid Feedforward Inhibition and Asynchronous Excitation Regulate Granule Cell Activity in the Mammalian Main Olfactory Bulb

PubMed Central

Burton, Shawn D.

2015-01-01

Granule cell-mediated inhibition is critical to patterning principal neuron activity in the olfactory bulb, and perturbation of synaptic input to granule cells significantly alters olfactory-guided behavior. Despite the critical role of granule cells in olfaction, little is known about how sensory input recruits granule cells. Here, we combined whole-cell patch-clamp electrophysiology in acute mouse olfactory bulb slices with biophysical multicompartmental modeling to investigate the synaptic basis of granule cell recruitment. Physiological activation of sensory afferents within single glomeruli evoked diverse modes of granule cell activity, including subthreshold depolarization, spikelets, and suprathreshold responses with widely distributed spike latencies. The generation of these diverse activity modes depended, in part, on the asynchronous time course of synaptic excitation onto granule cells, which lasted several hundred milliseconds. In addition to asynchronous excitation, each granule cell also received synchronous feedforward inhibition. This inhibition targeted both proximal somatodendritic and distal apical dendritic domains of granule cells, was reliably recruited across sniff rhythms, and scaled in strength with excitation as more glomeruli were activated. Feedforward inhibition onto granule cells originated from deep short-axon cells, which responded to glomerular activation with highly reliable, short-latency firing consistent with tufted cell-mediated excitation. Simulations showed that feedforward inhibition interacts with asynchronous excitation to broaden granule cell spike latency distributions and significantly attenuates granule cell depolarization within local subcellular compartments. Collectively, our results thus identify feedforward inhibition onto granule cells as a core feature of olfactory bulb circuitry and establish asynchronous excitation and feedforward inhibition as critical regulators of granule cell activity. SIGNIFICANCE STATEMENT Inhibitory granule cells are involved critically in shaping odor-evoked principal neuron activity in the mammalian olfactory bulb, yet little is known about how sensory input activates granule cells. Here, we show that sensory input to the olfactory bulb evokes a barrage of asynchronous synaptic excitation and highly reliable, short-latency synaptic inhibition onto granule cells via a disynaptic feedforward inhibitory circuit involving deep short-axon cells. Feedforward inhibition attenuates local depolarization within granule cell dendritic branches, interacts with asynchronous excitation to suppress granule cell spike-timing precision, and scales in strength with excitation across different levels of sensory input to normalize granule cell firing rates. PMID:26490853

Neural network approach to the inverse problem of the crack-depth determination from ultrasonic backscattering data

NASA Astrophysics Data System (ADS)

Takadoya, M.; Notake, M.; Kitahara, M.; Achenbach, J. D.; Guo, Q. C.; Peterson, M. L.

A neural network approach has been developed to determine the depth of a surface breaking crack in a steel plate from ultrasonic backscattering data. The network is trained by the use of a feedforward three-layered network together with a back-propagation algorithm for error corrections. Synthetic data are employed for network training. The signal used for crack isonification is a mode converted 45 deg transverse wave. The plate with a surface breaking crack is immersed in water, and the crack is insonified from the opposite uncracked side of the plate. A numerical analysis of the backscattered field is carried out based on the elastic wave theory by the use of the boundary element method. The numerical analysis provides synthetic data for the training of the network. The training data have been calculated for cracks with specific increments of the experimental data which are different from the training data.

In silico prediction of nematic transition temperature for liquid crystals using quantitative structure-property relationship approaches.

PubMed

Fatemi, Mohammad Hossein; Ghorbanzad'e, Mehdi

2009-11-01

Quantitative structure-property relationship models for the prediction of the nematic transition temperature (T (N)) were developed by using multilinear regression analysis and a feedforward artificial neural network (ANN). A collection of 42 thermotropic liquid crystals was chosen as the data set. The data set was divided into three sets: for training, and an internal and external test set. Training and internal test sets were used for ANN model development, and the external test set was used for evaluation of the predictive power of the model. In order to build the models, a set of six descriptors were selected by the best multilinear regression procedure of the CODESSA program. These descriptors were: atomic charge weighted partial negatively charged surface area, relative negative charged surface area, polarity parameter/square distance, minimum most negative atomic partial charge, molecular volume, and the A component of moment of inertia, which encode geometrical and electronic characteristics of molecules. These descriptors were used as inputs to ANN. The optimized ANN model had 6:6:1 topology. The standard errors in the calculation of T (N) for the training, internal, and external test sets using the ANN model were 1.012, 4.910, and 4.070, respectively. To further evaluate the ANN model, a crossvalidation test was performed, which produced the statistic Q (2) = 0.9796 and standard deviation of 2.67 based on predicted residual sum of square. Also, the diversity test was performed to ensure the model's stability and prove its predictive capability. The obtained results reveal the suitability of ANN for the prediction of T (N) for liquid crystals using molecular structural descriptors.

Multiple-Input Subject-Specific Modeling of Plasma Glucose Concentration for Feedforward Control.

PubMed

Kotz, Kaylee; Cinar, Ali; Mei, Yong; Roggendorf, Amy; Littlejohn, Elizabeth; Quinn, Laurie; Rollins, Derrick K

2014-11-26

The ability to accurately develop subject-specific, input causation models, for blood glucose concentration (BGC) for large input sets can have a significant impact on tightening control for insulin dependent diabetes. More specifically, for Type 1 diabetics (T1Ds), it can lead to an effective artificial pancreas (i.e., an automatic control system that delivers exogenous insulin) under extreme changes in critical disturbances. These disturbances include food consumption, activity variations, and physiological stress changes. Thus, this paper presents a free-living, outpatient, multiple-input, modeling method for BGC with strong causation attributes that is stable and guards against overfitting to provide an effective modeling approach for feedforward control (FFC). This approach is a Wiener block-oriented methodology, which has unique attributes for meeting critical requirements for effective, long-term, FFC.

Feedforward responses of transversus abdominis are directionally specific and act asymmetrically: implications for core stability theories.

PubMed

Allison, Garry T; Morris, Sue L; Lay, Brendan

2008-05-01

Experimental laboratory study supplemented with a repeated case study. To examine bilateral muscle activity of the deep abdominals in response to rapid arm raising, specifically to examine the laterality and directional specificity of feedforward responses of the transversus abdominis (TrA). Based on the feedforward responses of trunk muscles during rapid arm movements, authors have concluded that the deep trunk muscles have different control mechanisms compared to the more superficial muscles. It has been proposed that deep trunk muscles such as TrA contribute substantially to the stability of the lumbar spine and that this is achieved through simultaneous bilateral feedforward activation. These inferences are based on unilateral fine-wire electromyographic (EMG) data and there are limited investigations of bilateral responses of the TrA during unilateral arm raising. Bilateral fine-wire and surface EMG data from the anterior deltoid, TrA, obliquus internus (OI), obliquus externus, biceps femoris, erector spinae, and rectus abdominis during repeated arm raises were recorded at 2 kHz. EMG signal linear envelopes were synchronized to the onset of the anterior deltoid. A feedforward window was defined as the period up to 50 ms after the onset of the anterior deltoid, and paired onsets for bilateral muscles were plotted for both left and right arm movements. Trunk muscles from the group data demonstrated differences between sides (laterality), which were systematically altered when alternate arms were raised (directional specificity). This was clearly evident for the TrA but less obvious for the erector spinae. The ipsilateral biceps femoris and obliquus externus, and contralateral OI and TrA, were activated earlier than the alternate side for both right and left arm movements. This was a consistent pattern over a 7-year period for the case study. Data for the rectus abdominis derived from the case study demonstrated little laterality or directionally specific response. This is the first study to show that the feedforward activity of the TrA is specific to the direction of arm movement and not bilaterally symmetrical. The asymmetry of TrA activity during arm raising suggests that the interpretation of the role of TrA as a bilateral stabilizer during anticipatory postural adjustments needs to be revised. Future research needs to examine muscle synergies associated with the asymmetrical function of the TrA and the underlying mechanism associated with low-load stability training. Therapy, level 5.

Integrating a piezoelectric actuator with mechanical and hydraulic devices to control camless engines

NASA Astrophysics Data System (ADS)

Mercorelli, Paolo; Werner, Nils

2016-10-01

The paper deals with some interdisciplinary aspects and problems concerning the actuation control which occur in the integration of a piezoelectric structure in an aggregate actuator consisting of a piezoelectric, a stroke ratio displacement, a mechanical and a hydraulic part. Problems like compensation of the piezo hysteresis effect, scaling force-position to obtain an adequate displacement of the actuator and finally the control of such a complex aggregate system are considered and solved. Even though this work considers a particular application, the solutions proposed in the paper are quite general. In fact, the considered technical aspects occurring in systems which utilize piezoelectric technologies can be used in a variegated gamma of actuators integrating piezoelectric technologies. A cascade controller is proposed to combine a Feedforward action with an internal and an external PI-Controller. The Feedforward Controller is based on the model of the whole actuator, so particular attention is paid to the model structure. The resulting Feedforward action is an adaptive one to compensate hydraulic pressure faults. Real measurements are shown.

Study of a Simulation Tool to Determine Achievable Control Dynamics and Control Power Requirements with Perfect Tracking

NASA Technical Reports Server (NTRS)

Ostroff, Aaron J.

1998-01-01

This paper contains a study of two methods for use in a generic nonlinear simulation tool that could be used to determine achievable control dynamics and control power requirements while performing perfect tracking maneuvers over the entire flight envelope. The two methods are NDI (nonlinear dynamic inversion) and the SOFFT(Stochastic Optimal Feedforward and Feedback Technology) feedforward control structure. Equivalent discrete and continuous SOFFT feedforward controllers have been developed. These equivalent forms clearly show that the closed-loop plant model loop is a plant inversion and is the same as the NDI formulation. The main difference is that the NDI formulation has a closed-loop controller structure whereas SOFFT uses an open-loop command model. Continuous, discrete, and hybrid controller structures have been developed and integrated into the formulation. Linear simulation results show that seven different configurations all give essentially the same response, with the NDI hybrid being slightly different. The SOFFT controller gave better tracking performance compared to the NDI controller when a nonlinear saturation element was added. Future plans include evaluation using a nonlinear simulation.

Active feedforward noise control and signal tracking of headsets: Electroacoustic analysis and system implementation.

PubMed

Bai, Mingsian R; Pan, Weichi; Chen, Hungyu

2018-03-01

Active noise control (ANC) of headsets is revisited in this paper. An in-depth electroacoustic analysis of the combined loudspeaker-cavity headset system is conducted on the basis of electro-mechano-acoustical analogous circuits. Model matching of the primary path and the secondary path leads to a feedforward control architecture. The ideal controller sheds some light on the key parameters that affect the noise reduction performance. Filtered-X least-mean-squares algorithm is employed to implement the feedforward controller on a digital signal processor. Since the relative delay of the primary path and the secondary path is crucial to the noise reduction performance, multirate signal processing with polyphase implementation is utilized to minimize the effective analog-digital conversion delay in the secondary path. Ad hoc decimation and interpolation filters are designed in order not to introduce excessive phase delays at the cutoff. Real-time experiments are undertaken to validate the implemented ANC system. Listening tests are also conducted to compare the fixed controller and the adaptive controller in terms of noise reduction and signal tracking performance for three noise types. The results have demonstrated that the fixed feedforward controller achieved satisfactory noise reduction performance and signal tracking quality.

Meta-modeling of the pesticide fate model MACRO for groundwater exposure assessments using artificial neural networks

NASA Astrophysics Data System (ADS)

Stenemo, Fredrik; Lindahl, Anna M. L.; Gärdenäs, Annemieke; Jarvis, Nicholas

2007-08-01

Several simple index methods that use easily accessible data have been developed and included in decision-support systems to estimate pesticide leaching across larger areas. However, these methods often lack important process descriptions (e.g. macropore flow), which brings into question their reliability. Descriptions of macropore flow have been included in simulation models, but these are too complex and demanding for spatial applications. To resolve this dilemma, a neural network simulation meta-model of the dual-permeability macropore flow model MACRO was created for pesticide groundwater exposure assessment. The model was parameterized using pedotransfer functions that require as input the clay and sand content of the topsoil and subsoil, and the topsoil organic carbon content. The meta-model also requires the topsoil pesticide half-life and the soil organic carbon sorption coefficient as input. A fully connected feed-forward multilayer perceptron classification network with two hidden layers, linked to fully connected feed-forward multilayer perceptron neural networks with one hidden layer, trained on sub-sets of the target variable, was shown to be a suitable meta-model for the intended purpose. A Fourier amplitude sensitivity test showed that the model output (the 80th percentile average yearly pesticide concentration at 1 m depth for a 20 year simulation period) was sensitive to all input parameters. The two input parameters related to pesticide characteristics (i.e. soil organic carbon sorption coefficient and topsoil pesticide half-life) were the most influential, but texture in the topsoil was also quite important since it was assumed to control the mass exchange coefficient that regulates the strength of macropore flow. This is in contrast to models based on the advection-dispersion equation where soil texture is relatively unimportant. The use of the meta-model is exemplified with a case-study where the spatial variability of pesticide leaching is mapped for a small field. It was shown that the area of the field that contributes most to leaching depends on the properties of the compound in question. It is concluded that the simulation meta-model of MACRO should prove useful for mapping relative pesticide leaching risks at large scales.

Feedforward-Feedback Hybrid Control for Magnetic Shape Memory Alloy Actuators Based on the Krasnosel'skii-Pokrovskii Model

PubMed Central

Zhou, Miaolei; Zhang, Qi; Wang, Jingyuan

2014-01-01

As a new type of smart material, magnetic shape memory alloy has the advantages of a fast response frequency and outstanding strain capability in the field of microdrive and microposition actuators. The hysteresis nonlinearity in magnetic shape memory alloy actuators, however, limits system performance and further application. Here we propose a feedforward-feedback hybrid control method to improve control precision and mitigate the effects of the hysteresis nonlinearity of magnetic shape memory alloy actuators. First, hysteresis nonlinearity compensation for the magnetic shape memory alloy actuator is implemented by establishing a feedforward controller which is an inverse hysteresis model based on Krasnosel'skii-Pokrovskii operator. Secondly, the paper employs the classical Proportion Integration Differentiation feedback control with feedforward control to comprise the hybrid control system, and for further enhancing the adaptive performance of the system and improving the control accuracy, the Radial Basis Function neural network self-tuning Proportion Integration Differentiation feedback control replaces the classical Proportion Integration Differentiation feedback control. Utilizing self-learning ability of the Radial Basis Function neural network obtains Jacobian information of magnetic shape memory alloy actuator for the on-line adjustment of parameters in Proportion Integration Differentiation controller. Finally, simulation results show that the hybrid control method proposed in this paper can greatly improve the control precision of magnetic shape memory alloy actuator and the maximum tracking error is reduced from 1.1% in the open-loop system to 0.43% in the hybrid control system. PMID:24828010

Feedforward-feedback hybrid control for magnetic shape memory alloy actuators based on the Krasnosel'skii-Pokrovskii model.

PubMed

Zhou, Miaolei; Zhang, Qi; Wang, Jingyuan

2014-01-01

As a new type of smart material, magnetic shape memory alloy has the advantages of a fast response frequency and outstanding strain capability in the field of microdrive and microposition actuators. The hysteresis nonlinearity in magnetic shape memory alloy actuators, however, limits system performance and further application. Here we propose a feedforward-feedback hybrid control method to improve control precision and mitigate the effects of the hysteresis nonlinearity of magnetic shape memory alloy actuators. First, hysteresis nonlinearity compensation for the magnetic shape memory alloy actuator is implemented by establishing a feedforward controller which is an inverse hysteresis model based on Krasnosel'skii-Pokrovskii operator. Secondly, the paper employs the classical Proportion Integration Differentiation feedback control with feedforward control to comprise the hybrid control system, and for further enhancing the adaptive performance of the system and improving the control accuracy, the Radial Basis Function neural network self-tuning Proportion Integration Differentiation feedback control replaces the classical Proportion Integration Differentiation feedback control. Utilizing self-learning ability of the Radial Basis Function neural network obtains Jacobian information of magnetic shape memory alloy actuator for the on-line adjustment of parameters in Proportion Integration Differentiation controller. Finally, simulation results show that the hybrid control method proposed in this paper can greatly improve the control precision of magnetic shape memory alloy actuator and the maximum tracking error is reduced from 1.1% in the open-loop system to 0.43% in the hybrid control system.

Estimation of Subpixel Snow-Covered Area by Nonparametric Regression Splines

NASA Astrophysics Data System (ADS)

Kuter, S.; Akyürek, Z.; Weber, G.-W.

2016-10-01

Measurement of the areal extent of snow cover with high accuracy plays an important role in hydrological and climate modeling. Remotely-sensed data acquired by earth-observing satellites offer great advantages for timely monitoring of snow cover. However, the main obstacle is the tradeoff between temporal and spatial resolution of satellite imageries. Soft or subpixel classification of low or moderate resolution satellite images is a preferred technique to overcome this problem. The most frequently employed snow cover fraction methods applied on Moderate Resolution Imaging Spectroradiometer (MODIS) data have evolved from spectral unmixing and empirical Normalized Difference Snow Index (NDSI) methods to latest machine learning-based artificial neural networks (ANNs). This study demonstrates the implementation of subpixel snow-covered area estimation based on the state-of-the-art nonparametric spline regression method, namely, Multivariate Adaptive Regression Splines (MARS). MARS models were trained by using MODIS top of atmospheric reflectance values of bands 1-7 as predictor variables. Reference percentage snow cover maps were generated from higher spatial resolution Landsat ETM+ binary snow cover maps. A multilayer feed-forward ANN with one hidden layer trained with backpropagation was also employed to estimate the percentage snow-covered area on the same data set. The results indicated that the developed MARS model performed better than th

Development of a helicopter rotor/propulsion system dynamics analysis

NASA Technical Reports Server (NTRS)

Warmbrodt, W.; Hull, R.

1982-01-01

A time-domain analysis of coupled engine/drive train/rotor dynamics of a twin-engine, single main rotor helicopter model has been performed. The analysis incorporates an existing helicopter model with nonlinear simulations of a helicopter turboshaft engine and its fuel controller. System dynamic behavior is studied using the resulting simulation which included representations for the two engines and their fuel controllers, drive system, main rotor, tail rotor, and aircraft rigid body motions. Time histories of engine and rotor RPM response to pilot control inputs are studied for a baseline rotor and propulsion system model. Sensitivity of rotor RPM droop to fuel controller gain changes and collective input feed-forward gain changes are studied. Torque-load-sharing between the two engines is investigated by making changes in the fuel controller feedback paths. A linear engine model is derived from the nonlinear engine simulation and used in the coupled system analysis. This four-state linear engine model is then reduced to a three-state model. The effect of this simplification on coupled system behavior is shown.

Post-Movement Beta Activity in Sensorimotor Cortex Indexes Confidence in the Estimations from Internal Models.

PubMed

Tan, Huiling; Wade, Cian; Brown, Peter

2016-02-03

Beta oscillations are a dominant feature of the sensorimotor system. A transient and prominent increase in beta oscillations is consistently observed across the sensorimotor cortical-basal ganglia network after cessation of voluntary movement: the post-movement beta synchronization (PMBS). Current theories about the function of the PMBS have been focused on either the closure of motor response or the processing of sensory afferance. Computational models of sensorimotor control have emphasized the importance of the integration between feedforward estimation and sensory feedback, and therefore the putative motor and sensory functions of beta oscillations may reciprocally interact with each other and in fact be indissociable. Here we show that the amplitude of sensorimotor PMBS is modulated by the history of visual feedback of task-relevant errors, and negatively correlated with the trial-to-trial exploratory adjustment in a sensorimotor adaptation task in young healthy human subjects. The PMBS also negatively correlated with the uncertainty associated with the feedforward estimation, which was recursively updated in light of new sensory feedback, as identified by a Bayesian learning model. These results reconcile the two opposing motor and sensory views of the function of PMBS, and suggest a unifying theory in which PMBS indexes the confidence in internal feedforward estimation in Bayesian sensorimotor integration. Its amplitude simultaneously reflects cortical sensory processing and signals the need for maintenance or adaptation of the motor output, and if necessary, exploration to identify an altered sensorimotor transformation. For optimal sensorimotor control, sensory feedback and feedforward estimation of a movement's sensory consequences should be weighted by the inverse of their corresponding uncertainties, which require recursive updating in a dynamic environment. We show that post-movement beta activity (13-30 Hz) over sensorimotor cortex in young healthy subjects indexes the evaluation of uncertainty in feedforward estimation. Our work contributes to the understanding of the function of beta oscillations in sensorimotor control, and provides further insight into how aberrant beta activity can contribute to the pathophysiology of movement disorders. Copyright © 2016 Tan et al.

Analog hardware for learning neural networks

NASA Technical Reports Server (NTRS)

Eberhardt, Silvio P. (Inventor)

1991-01-01

This is a recurrent or feedforward analog neural network processor having a multi-level neuron array and a synaptic matrix for storing weighted analog values of synaptic connection strengths which is characterized by temporarily changing one connection strength at a time to determine its effect on system output relative to the desired target. That connection strength is then adjusted based on the effect, whereby the processor is taught the correct response to training examples connection by connection.

Input filter compensation for switching regulators

NASA Technical Reports Server (NTRS)

Kelkar, S. S.; Lee, F. C.

1983-01-01

A novel input filter compensation scheme for a buck regulator that eliminates the interaction between the input filter output impedance and the regulator control loop is presented. The scheme is implemented using a feedforward loop that senses the input filter state variables and uses this information to modulate the duty cycle signal. The feedforward design process presented is seen to be straightforward and the feedforward easy to implement. Extensive experimental data supported by analytical results show that significant performance improvement is achieved with the use of feedforward in the following performance categories: loop stability, audiosusceptibility, output impedance and transient response. The use of feedforward results in isolating the switching regulator from its power source thus eliminating all interaction between the regulator and equipment upstream. In addition the use of feedforward removes some of the input filter design constraints and makes the input filter design process simpler thus making it possible to optimize the input filter. The concept of feedforward compensation can also be extended to other types of switching regulators.

Deep Neural Networks: A New Framework for Modeling Biological Vision and Brain Information Processing.

PubMed

Kriegeskorte, Nikolaus

2015-11-24

Recent advances in neural network modeling have enabled major strides in computer vision and other artificial intelligence applications. Human-level visual recognition abilities are coming within reach of artificial systems. Artificial neural networks are inspired by the brain, and their computations could be implemented in biological neurons. Convolutional feedforward networks, which now dominate computer vision, take further inspiration from the architecture of the primate visual hierarchy. However, the current models are designed with engineering goals, not to model brain computations. Nevertheless, initial studies comparing internal representations between these models and primate brains find surprisingly similar representational spaces. With human-level performance no longer out of reach, we are entering an exciting new era, in which we will be able to build biologically faithful feedforward and recurrent computational models of how biological brains perform high-level feats of intelligence, including vision.

Modeling gene regulatory network motifs using statecharts

PubMed Central

2012-01-01

Background Gene regulatory networks are widely used by biologists to describe the interactions among genes, proteins and other components at the intra-cellular level. Recently, a great effort has been devoted to give gene regulatory networks a formal semantics based on existing computational frameworks. For this purpose, we consider Statecharts, which are a modular, hierarchical and executable formal model widely used to represent software systems. We use Statecharts for modeling small and recurring patterns of interactions in gene regulatory networks, called motifs. Results We present an improved method for modeling gene regulatory network motifs using Statecharts and we describe the successful modeling of several motifs, including those which could not be modeled or whose models could not be distinguished using the method of a previous proposal. We model motifs in an easy and intuitive way by taking advantage of the visual features of Statecharts. Our modeling approach is able to simulate some interesting temporal properties of gene regulatory network motifs: the delay in the activation and the deactivation of the "output" gene in the coherent type-1 feedforward loop, the pulse in the incoherent type-1 feedforward loop, the bistability nature of double positive and double negative feedback loops, the oscillatory behavior of the negative feedback loop, and the "lock-in" effect of positive autoregulation. Conclusions We present a Statecharts-based approach for the modeling of gene regulatory network motifs in biological systems. The basic motifs used to build more complex networks (that is, simple regulation, reciprocal regulation, feedback loop, feedforward loop, and autoregulation) can be faithfully described and their temporal dynamics can be analyzed. PMID:22536967

Activation of response force by self-splitting objects: where are the limits of feedforward Gestalt processing?

PubMed

Schmidt, Filipp; Weber, Andreas; Schmidt, Thomas

2014-08-21

Most objects can be recognized easily even when they are partly occluded. This also holds when several overlapping objects share the same surface features (self-splitting objects) which is an illustration of the grouping principle of Good Gestalt. We employed outline and filled contour stimuli in a primed flanker task to test whether the processing of self-splitting objects is in accordance with a simple feedforward model. We obtained priming effects in response time and response force for both types of stimuli, even when increasing the number of occluders up to three. The results for outline contours were in full accordance with a feedforward account. This was not the case for the results for filled contours (i.e., for self-splitting objects), especially under conditions of strong occlusion. We conclude that the implementation of the Good Gestalt principle is fast but still based on recurrent processing. © 2014 ARVO.

Numerical solution of differential equations by artificial neural networks

NASA Technical Reports Server (NTRS)

Meade, Andrew J., Jr.

1995-01-01

Conventionally programmed digital computers can process numbers with great speed and precision, but do not easily recognize patterns or imprecise or contradictory data. Instead of being programmed in the conventional sense, artificial neural networks (ANN's) are capable of self-learning through exposure to repeated examples. However, the training of an ANN can be a time consuming and unpredictable process. A general method is being developed by the author to mate the adaptability of the ANN with the speed and precision of the digital computer. This method has been successful in building feedforward networks that can approximate functions and their partial derivatives from examples in a single iteration. The general method also allows the formation of feedforward networks that can approximate the solution to nonlinear ordinary and partial differential equations to desired accuracy without the need of examples. It is believed that continued research will produce artificial neural networks that can be used with confidence in practical scientific computing and engineering applications.

High pressure air compressor valve fault diagnosis using feedforward neural networks

NASA Astrophysics Data System (ADS)

James Li, C.; Yu, Xueli

1995-09-01

Feedforward neural networks (FNNs) are developed and implemented to classify a four-stage high pressure air compressor into one of the following conditions: baseline, suction or exhaust valve faults. These FNNs are used for the compressor's automatic condition monitoring and fault diagnosis. Measurements of 39 variables are obtained under different baseline conditions and third-stage suction and exhaust valve faults. These variables include pressures and temperatures at all stages, voltage between phase aand phase b, voltage between phase band phase c, total three-phase real power, cooling water flow rate, etc. To reduce the number of variables, the amount of their discriminatory information is quantified by scattering matrices to identify statistical significant ones. Measurements of the selected variables are then used by a fully automatic structural and weight learning algorithm to construct three-layer FNNs to classify the compressor's condition. This learning algorithm requires neither guesses of initial weight values nor number of neurons in the hidden layer of an FNN. It takes an incremental approach in which a hidden neuron is trained by exemplars and then augmented to the existing network. These exemplars are then made orthogonal to the newly identified hidden neuron. They are subsequently used for the training of the next hidden neuron. The betterment continues until a desired accuracy is reached. After the neural networks are established, novel measurements from various conditions that haven't been previously seen by the FNNs are then used to evaluate their ability in fault diagnosis. The trained neural networks provide very accurate diagnosis for suction and discharge valve defects.

Modeling and simulation of xylitol production in bioreactor by Debaryomyces nepalensis NCYC 3413 using unstructured and artificial neural network models.

PubMed

Pappu, J Sharon Mano; Gummadi, Sathyanarayana N

2016-11-01

This study examines the use of unstructured kinetic model and artificial neural networks as predictive tools for xylitol production by Debaryomyces nepalensis NCYC 3413 in bioreactor. An unstructured kinetic model was proposed in order to assess the influence of pH (4, 5 and 6), temperature (25°C, 30°C and 35°C) and volumetric oxygen transfer coefficient kLa (0.14h(-1), 0.28h(-1) and 0.56h(-1)) on growth and xylitol production. A feed-forward back-propagation artificial neural network (ANN) has been developed to investigate the effect of process condition on xylitol production. ANN configuration of 6-10-3 layers was selected and trained with 339 experimental data points from bioreactor studies. Results showed that simulation and prediction accuracy of ANN was apparently higher when compared to unstructured mechanistic model under varying operational conditions. ANN was found to be an efficient data-driven tool to predict the optimal harvest time in xylitol production. Copyright © 2016 Elsevier Ltd. All rights reserved.

Estimation of biogas and methane yields in an UASB treating potato starch processing wastewater with backpropagation artificial neural network.

PubMed

Antwi, Philip; Li, Jianzheng; Boadi, Portia Opoku; Meng, Jia; Shi, En; Deng, Kaiwen; Bondinuba, Francis Kwesi

2017-03-01

Three-layered feedforward backpropagation (BP) artificial neural networks (ANN) and multiple nonlinear regression (MnLR) models were developed to estimate biogas and methane yield in an upflow anaerobic sludge blanket (UASB) reactor treating potato starch processing wastewater (PSPW). Anaerobic process parameters were optimized to identify their importance on methanation. pH, total chemical oxygen demand, ammonium, alkalinity, total Kjeldahl nitrogen, total phosphorus, volatile fatty acids and hydraulic retention time selected based on principal component analysis were used as input variables, whiles biogas and methane yield were employed as target variables. Quasi-Newton method and conjugate gradient backpropagation algorithms were best among eleven training algorithms. Coefficient of determination (R 2 ) of the BP-ANN reached 98.72% and 97.93% whiles MnLR model attained 93.9% and 91.08% for biogas and methane yield, respectively. Compared with the MnLR model, BP-ANN model demonstrated significant performance, suggesting possible control of the anaerobic digestion process with the BP-ANN model. Copyright © 2016 Elsevier Ltd. All rights reserved.

Aircraft automatic-flight-control system with inversion of the model in the feed-forward path using a Newton-Raphson technique for the inversion

NASA Technical Reports Server (NTRS)

Smith, G. A.; Meyer, G.; Nordstrom, M.

1986-01-01

A new automatic flight control system concept suitable for aircraft with highly nonlinear aerodynamic and propulsion characteristics and which must operate over a wide flight envelope was investigated. This exact model follower inverts a complete nonlinear model of the aircraft as part of the feed-forward path. The inversion is accomplished by a Newton-Raphson trim of the model at each digital computer cycle time of 0.05 seconds. The combination of the inverse model and the actual aircraft in the feed-forward path alloys the translational and rotational regulators in the feedback path to be easily designed by linear methods. An explanation of the model inversion procedure is presented. An extensive set of simulation data for essentially the full flight envelope for a vertical attitude takeoff and landing aircraft (VATOL) is presented. These data demonstrate the successful, smooth, and precise control that can be achieved with this concept. The trajectory includes conventional flight from 200 to 900 ft/sec with path accelerations and decelerations, altitude changes of over 6000 ft and 2g and 3g turns. Vertical attitude maneuvering as a tail sitter along all axes is demonstrated. A transition trajectory from 200 ft/sec in conventional flight to stationary hover in the vertical attitude includes satisfactory operation through lift-cure slope reversal as attitude goes from horizontal to vertical at constant altitude. A vertical attitude takeoff from stationary hover to conventional flight is also demonstrated.

An adaptive Cartesian control scheme for manipulators

NASA Technical Reports Server (NTRS)

Seraji, H.

1987-01-01

A adaptive control scheme for direct control of manipulator end-effectors to achieve trajectory tracking in Cartesian space is developed. The control structure is obtained from linear multivariable theory and is composed of simple feedforward and feedback controllers and an auxiliary input. The direct adaptation laws are derived from model reference adaptive control theory and are not based on parameter estimation of the robot model. The utilization of feedforward control and the inclusion of auxiliary input are novel features of the present scheme and result in improved dynamic performance over existing adaptive control schemes. The adaptive controller does not require the complex mathematical model of the robot dynamics or any knowledge of the robot parameters or the payload, and is computationally fast for online implementation with high sampling rates.

Artificial neural systems for interpretation and inversion of seismic data

NASA Astrophysics Data System (ADS)

Calderon-Macias, Carlos

The goal of this work is to investigate the feasibility of using neural network (NN) models for solving geophysical exploration problems. First, a feedforward neural network (FNN) is used to solve inverse problems. The operational characteristics of a FNN are primarily controlled by a set of weights and a nonlinear function that performs a mapping between two sets of data. In a process known as training, the FNN weights are iteratively adjusted to perform the mapping. After training, the computed weights encode important features of the data that enable one pattern to be distinguished from another. Synthetic data computed from an ensemble of earth models and the corresponding models provide the training data. Two training methods are studied: the backpropagation method which is a gradient scheme, and a global optimization method called very fast simulated annealing (VFSA). A trained network is then used to predict models from new data (e.g., data from a new location) in a one-step procedure. The application of this method to the problems of obtaining formation resistivities and layer thicknesses from resistivity sounding data and 1D velocity models from seismic data shows that trained FNNs produce reasonably accurate earth models when observed data are input to the FNNs. In a second application, a FNN is used for automating the NMO correction process of seismic reflection data. The task of the FNN is to map CMP data at control locations along a seismic line into subsurface velocities. The network is trained while the velocity analyses are performed at the control locations. Once trained, the computed weights are used as an operator that acts on the remaining CMP data as a velocity interpolator, resulting in a fast method for NMO correction. The second part of this dissertation describes the application of a Hopfield neural network (HNN) to the problems of deconvolution and multiple attenuation. In these applications, the unknown parameters (reflection coefficients and source wavelet in the first problem and an operator in the second) are mapped as neurons of the HNN. The proposed deconvolution method attempts to reproduce the data with a limited number of events. The multiple attenuation method resembles the predictive deconvolution method. Results of this method are compared with a multiple elimination method based on estimating the source wavelet from the seismic data.

Ultra-precise tracking control of piezoelectric actuators via a fuzzy hysteresis model.

PubMed

Li, Pengzhi; Yan, Feng; Ge, Chuan; Zhang, Mingchao

2012-08-01

In this paper, a novel Takagi-Sugeno (T-S) fuzzy system based model is proposed for hysteresis in piezoelectric actuators. The antecedent and consequent structures of the fuzzy hysteresis model (FHM) can be, respectively, identified on-line through uniform partition approach and recursive least squares (RLS) algorithm. With respect to controller design, the inverse of FHM is used to develop a feedforward controller to cancel out the hysteresis effect. Then a hybrid controller is designed for high-performance tracking. It combines the feedforward controller with a proportional integral differential (PID) controller favourable for stabilization and disturbance compensation. To achieve nanometer-scale tracking precision, the enhanced adaptive hybrid controller is further developed. It uses real-time input and output data to update FHM, thus changing the feedforward controller to suit the on-site hysteresis character of the piezoelectric actuator. Finally, as to 3 cases of 50 Hz sinusoidal, multiple frequency sinusoidal and 50 Hz triangular trajectories tracking, experimental results demonstrate the efficiency of the proposed controllers. Especially, being only 0.35% of the maximum desired displacement, the maximum error of 50 Hz sinusoidal tracking is greatly reduced to 5.8 nm, which clearly shows the ultra-precise nanometer-scale tracking performance of the developed adaptive hybrid controller.

Results of adaptive feedforward on GTA

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

Ziomek, C.D.; Denney, P.M.; Regan, A.H.

1993-01-01

This paper presents the results of the adaptive feedforward system in use on the Ground Test Accelerator (GTA). The adaptive feedforward system was shown to correct repetitive, high-frequency errors in the amplitude and phase of the RF field of the pulsed accelerator. The adaptive feedforward system was designed as an augmentation to the RF field feedback control system and was able to extend the closed-loop bandwidth and disturbance rejection by a factor of ten. Within a second implementation, the adaptive feedforward hardware was implemented in place of the feedback control system and was shown to negate both beam transients andmore » phase droop in the klystron amplifier.« less

Results of adaptive feedforward on GTA

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

Ziomek, C.D.; Denney, P.M.; Regan, A.H.

1993-06-01

This paper presents the results of the adaptive feedforward system in use on the Ground Test Accelerator (GTA). The adaptive feedforward system was shown to correct repetitive, high-frequency errors in the amplitude and phase of the RF field of the pulsed accelerator. The adaptive feedforward system was designed as an augmentation to the RF field feedback control system and was able to extend the closed-loop bandwidth and disturbance rejection by a factor of ten. Within a second implementation, the adaptive feedforward hardware was implemented in place of the feedback control system and was shown to negate both beam transients andmore » phase droop in the klystron amplifier.« less

Quantifying feedforward control: a linear scaling model for fingertip forces and object weight.

PubMed

Lu, Ying; Bilaloglu, Seda; Aluru, Viswanath; Raghavan, Preeti

2015-07-01

The ability to predict the optimal fingertip forces according to object properties before the object is lifted is known as feedforward control, and it is thought to occur due to the formation of internal representations of the object's properties. The control of fingertip forces to objects of different weights has been studied extensively by using a custom-made grip device instrumented with force sensors. Feedforward control is measured by the rate of change of the vertical (load) force before the object is lifted. However, the precise relationship between the rate of change of load force and object weight and how it varies across healthy individuals in a population is not clearly understood. Using sets of 10 different weights, we have shown that there is a log-linear relationship between the fingertip load force rates and weight among neurologically intact individuals. We found that after one practice lift, as the weight increased, the peak load force rate (PLFR) increased by a fixed percentage, and this proportionality was common among the healthy subjects. However, at any given weight, the level of PLFR varied across individuals and was related to the efficiency of the muscles involved in lifting the object, in this case the wrist and finger extensor muscles. These results quantify feedforward control during grasp and lift among healthy individuals and provide new benchmarks to interpret data from neurologically impaired populations as well as a means to assess the effect of interventions on restoration of feedforward control and its relationship to muscular control. Copyright © 2015 the American Physiological Society.

Biomechanical constraints on the feedforward regulation of endpoint stiffness.

PubMed

Hu, Xiao; Murray, Wendy M; Perreault, Eric J

2012-10-01

Although many daily tasks tend to destabilize arm posture, it is still possible to have stable interactions with the environment by regulating the multijoint mechanics of the arm in a task-appropriate manner. For postural tasks, this regulation involves the appropriate control of endpoint stiffness, which represents the stiffness of the arm at the hand. Although experimental studies have been used to evaluate endpoint stiffness control, including the orientation of maximal stiffness, the underlying neural strategies remain unknown. Specifically, the relative importance of feedforward and feedback mechanisms has yet to be determined due to the difficulty separately identifying the contributions of these mechanisms in human experiments. This study used a previously validated three-dimensional musculoskeletal model of the arm to quantify the degree to which the orientation of maximal endpoint stiffness could be changed using only steady-state muscle activations, used to represent feedforward motor commands. Our hypothesis was that the feedforward control of endpoint stiffness orientation would be significantly constrained by the biomechanical properties of the musculoskeletal system. Our results supported this hypothesis, demonstrating substantial biomechanical constraints on the ability to regulate endpoint stiffness throughout the workspace. The ability to regulate stiffness orientation was further constrained by additional task requirements, such as the need to support the arm against gravity or exert forces on the environment. Together, these results bound the degree to which slowly varying feedforward motor commands can be used to regulate the orientation of maximum arm stiffness and provide a context for better understanding conditions in which feedback control may be needed.

Automatic threshold selection for multi-class open set recognition

NASA Astrophysics Data System (ADS)

Scherreik, Matthew; Rigling, Brian

2017-05-01

Multi-class open set recognition is the problem of supervised classification with additional unknown classes encountered after a model has been trained. An open set classifer often has two core components. The first component is a base classifier which estimates the most likely class of a given example. The second component consists of open set logic which estimates if the example is truly a member of the candidate class. Such a system is operated in a feed-forward fashion. That is, a candidate label is first estimated by the base classifier, and the true membership of the example to the candidate class is estimated afterward. Previous works have developed an iterative threshold selection algorithm for rejecting examples from classes which were not present at training time. In those studies, a Platt-calibrated SVM was used as the base classifier, and the thresholds were applied to class posterior probabilities for rejection. In this work, we investigate the effectiveness of other base classifiers when paired with the threshold selection algorithm and compare their performance with the original SVM solution.

Emergence of network structure due to spike-timing-dependent plasticity in recurrent neuronal networks IV: structuring synaptic pathways among recurrent connections.

PubMed

Gilson, Matthieu; Burkitt, Anthony N; Grayden, David B; Thomas, Doreen A; van Hemmen, J Leo

2009-12-01

In neuronal networks, the changes of synaptic strength (or weight) performed by spike-timing-dependent plasticity (STDP) are hypothesized to give rise to functional network structure. This article investigates how this phenomenon occurs for the excitatory recurrent connections of a network with fixed input weights that is stimulated by external spike trains. We develop a theoretical framework based on the Poisson neuron model to analyze the interplay between the neuronal activity (firing rates and the spike-time correlations) and the learning dynamics, when the network is stimulated by correlated pools of homogeneous Poisson spike trains. STDP can lead to both a stabilization of all the neuron firing rates (homeostatic equilibrium) and a robust weight specialization. The pattern of specialization for the recurrent weights is determined by a relationship between the input firing-rate and correlation structures, the network topology, the STDP parameters and the synaptic response properties. We find conditions for feed-forward pathways or areas with strengthened self-feedback to emerge in an initially homogeneous recurrent network.

Reinforcement Learning of Linking and Tracing Contours in Recurrent Neural Networks

PubMed Central

Brosch, Tobias; Neumann, Heiko; Roelfsema, Pieter R.

2015-01-01

The processing of a visual stimulus can be subdivided into a number of stages. Upon stimulus presentation there is an early phase of feedforward processing where the visual information is propagated from lower to higher visual areas for the extraction of basic and complex stimulus features. This is followed by a later phase where horizontal connections within areas and feedback connections from higher areas back to lower areas come into play. In this later phase, image elements that are behaviorally relevant are grouped by Gestalt grouping rules and are labeled in the cortex with enhanced neuronal activity (object-based attention in psychology). Recent neurophysiological studies revealed that reward-based learning influences these recurrent grouping processes, but it is not well understood how rewards train recurrent circuits for perceptual organization. This paper examines the mechanisms for reward-based learning of new grouping rules. We derive a learning rule that can explain how rewards influence the information flow through feedforward, horizontal and feedback connections. We illustrate the efficiency with two tasks that have been used to study the neuronal correlates of perceptual organization in early visual cortex. The first task is called contour-integration and demands the integration of collinear contour elements into an elongated curve. We show how reward-based learning causes an enhancement of the representation of the to-be-grouped elements at early levels of a recurrent neural network, just as is observed in the visual cortex of monkeys. The second task is curve-tracing where the aim is to determine the endpoint of an elongated curve composed of connected image elements. If trained with the new learning rule, neural networks learn to propagate enhanced activity over the curve, in accordance with neurophysiological data. We close the paper with a number of model predictions that can be tested in future neurophysiological and computational studies. PMID:26496502

Parallel regulation of feedforward inhibition and excitation during whisker map plasticity

PubMed Central

House, David RC; Elstrott, Justin; Koh, Eileen; Chung, Jason; Feldman, Daniel E.

2011-01-01

Sensory experience drives robust plasticity of sensory maps in cerebral cortex, but the role of inhibitory circuits in this process is not fully understood. We show that classical deprivation-induced whisker map plasticity in layer 2/3 (L2/3) of rat somatosensory (S1) cortex involves robust weakening of L4-L2/3 feedforward inhibition. This weakening was caused by reduced L4 excitation onto L2/3 fast-spiking (FS) interneurons, which mediate sensitive feedforward inhibition, and was partially offset by strengthening of unitary FS to L2/3 pyramidal cell synapses. Weakening of feedforward inhibition paralleled the known weakening of feedforward excitation, so that mean excitatory-inhibitory balance and timing onto L2/3 pyramidal cells were preserved. Thus, reduced feedforward inhibition is a covert compensatory process that can maintain excitatory-inhibitory balance during classical deprivation-induced Hebbian map plasticity. PMID:22153377

[Modeling developmental aspects of sensorimotor control of speech production].

PubMed

Kröger, B J; Birkholz, P; Neuschaefer-Rube, C

2007-05-01

Detailed knowledge of the neurophysiology of speech acquisition is important for understanding the developmental aspects of speech perception and production and for understanding developmental disorders of speech perception and production. A computer implemented neural model of sensorimotor control of speech production was developed. The model is capable of demonstrating the neural functions of different cortical areas during speech production in detail. (i) Two sensory and two motor maps or neural representations and the appertaining neural mappings or projections establish the sensorimotor feedback control system. These maps and mappings are already formed and trained during the prelinguistic phase of speech acquisition. (ii) The feedforward sensorimotor control system comprises the lexical map (representations of sounds, syllables, and words of the first language) and the mappings from lexical to sensory and to motor maps. The training of the appertaining mappings form the linguistic phase of speech acquisition. (iii) Three prelinguistic learning phases--i. e. silent mouthing, quasi stationary vocalic articulation, and realisation of articulatory protogestures--can be defined on the basis of our simulation studies using the computational neural model. These learning phases can be associated with temporal phases of prelinguistic speech acquisition obtained from natural data. The neural model illuminates the detailed function of specific cortical areas during speech production. In particular it can be shown that developmental disorders of speech production may result from a delayed or incorrect process within one of the prelinguistic learning phases defined by the neural model.

Influences of load characteristics on impaired control of grip forces in patients with cerebellar damage.

PubMed

Brandauer, B; Timmann, D; Häusler, A; Hermsdörfer, J

2010-02-01

Various studies showed a clear impairment of cerebellar patients to modulate grip force in anticipation of the loads resulting from movements with a grasped object. This failure corroborated the theory of internal feedforward models in the cerebellum. Cerebellar damage also impairs the coordination of multiple-joint movements and this has been related to deficient prediction and compensation of movement-induced torques. To study the effects of disturbed torque control on feedforward grip-force control, two self-generated load conditions with different demands on torque control-one with movement-induced and the other with isometrically generated load changes-were directly compared in patients with cerebellar degeneration. Furthermore the cerebellum is thought to be more involved in grip-force adjustment to self-generated loads than to externally generated loads. Consequently, an additional condition with externally generated loads was introduced to further test this hypothesis. Analysis of 23 patients with degenerative cerebellar damage revealed clear impairments in predictive feedforward mechanisms in the control of both self-generated load types. Besides feedforward control, the cerebellar damage also affected more reactive responses when the externally generated load destabilized the grip, although this impairment may vary with the type of load as suggested by control experiments. The present findings provide further support that the cerebellum plays a major role in predictive control mechanisms. However, this impact of the cerebellum does not strongly depend on the nature of the load and the specific internal forward model. Contributions to reactive (grip force) control are not negligible, but seem to be dependent on the physical characteristics of an externally generated load.

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.

Feedforward motor control in developmental dyslexia and developmental coordination disorder: Does comorbidity matter?

PubMed

Cignetti, Fabien; Vaugoyeau, Marianne; Fontan, Aurelie; Jover, Marianne; Livet, Marie-Odile; Hugonenq, Catherine; Audic, Frédérique; Chabrol, Brigitte; Assaiante, Christine

2018-05-01

Feedforward and online controls are two facets of predictive motor control from internal models, which is suspected to be impaired in learning disorders. We examined whether the feedforward component is affected in children (8-12 years) with developmental dyslexia (DD) and/or with developmental coordination disorder (DCD) compared to typically developing (TD) children. Children underwent a bimanual unloading paradigm during which a load supported to one arm, the postural arm, was either unexpectedly unloaded by a computer or voluntary unloaded by the subject with the other arm. All children showed a better stabilization (lower flexion) of the postural arm and an earlier inhibition of the arm flexors during voluntary unloading, indicating anticipation of unloading. Between-group comparisons of kinematics and electromyographic activity of the postural arm revealed that the difference during voluntary unloading was between DD-DCD children and the other groups, with the former showing a delayed inhibition of the flexor muscles. Deficit of the feedforward component of motor control may particularly apply to comorbid subtypes, here the DD-DCD subtype. The development of a comprehensive framework for motor performance deficits in children with learning disorders will be achieved only by dissociating key components of motor prediction and focusing on subtypes and comorbidities. Copyright © 2018 Elsevier Ltd. All rights reserved.

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.

Understanding the role of speech production in reading: Evidence for a print-to-speech neural network using graphical analysis.

PubMed

Cummine, Jacqueline; Cribben, Ivor; Luu, Connie; Kim, Esther; Bahktiari, Reyhaneh; Georgiou, George; Boliek, Carol A

2016-05-01

The neural circuitry associated with language processing is complex and dynamic. Graphical models are useful for studying complex neural networks as this method provides information about unique connectivity between regions within the context of the entire network of interest. Here, the authors explored the neural networks during covert reading to determine the role of feedforward and feedback loops in covert speech production. Brain activity of skilled adult readers was assessed in real word and pseudoword reading tasks with functional MRI (fMRI). The authors provide evidence for activity coherence in the feedforward system (inferior frontal gyrus-supplementary motor area) during real word reading and in the feedback system (supramarginal gyrus-precentral gyrus) during pseudoword reading. Graphical models provided evidence of an extensive, highly connected, neural network when individuals read real words that relied on coordination of the feedforward system. In contrast, when individuals read pseudowords the authors found a limited/restricted network that relied on coordination of the feedback system. Together, these results underscore the importance of considering multiple pathways and articulatory loops during language tasks and provide evidence for a print-to-speech neural network. (PsycINFO Database Record (c) 2016 APA, all rights reserved).

Enhanced HMAX model with feedforward feature learning for multiclass categorization.

PubMed

Li, Yinlin; Wu, Wei; Zhang, Bo; Li, Fengfu

2015-01-01

In recent years, the interdisciplinary research between neuroscience and computer vision has promoted the development in both fields. Many biologically inspired visual models are proposed, and among them, the Hierarchical Max-pooling model (HMAX) is a feedforward model mimicking the structures and functions of V1 to posterior inferotemporal (PIT) layer of the primate visual cortex, which could generate a series of position- and scale- invariant features. However, it could be improved with attention modulation and memory processing, which are two important properties of the primate visual cortex. Thus, in this paper, based on recent biological research on the primate visual cortex, we still mimic the first 100-150 ms of visual cognition to enhance the HMAX model, which mainly focuses on the unsupervised feedforward feature learning process. The main modifications are as follows: (1) To mimic the attention modulation mechanism of V1 layer, a bottom-up saliency map is computed in the S1 layer of the HMAX model, which can support the initial feature extraction for memory processing; (2) To mimic the learning, clustering and short-term memory to long-term memory conversion abilities of V2 and IT, an unsupervised iterative clustering method is used to learn clusters with multiscale middle level patches, which are taken as long-term memory; (3) Inspired by the multiple feature encoding mode of the primate visual cortex, information including color, orientation, and spatial position are encoded in different layers of the HMAX model progressively. By adding a softmax layer at the top of the model, multiclass categorization experiments can be conducted, and the results on Caltech101 show that the enhanced model with a smaller memory size exhibits higher accuracy than the original HMAX model, and could also achieve better accuracy than other unsupervised feature learning methods in multiclass categorization task.

Incorporating geographical factors with artificial neural networks to predict reference values of erythrocyte sedimentation rate

PubMed Central

2013-01-01

Background The measurement of the Erythrocyte Sedimentation Rate (ESR) value is a standard procedure performed during a typical blood test. In order to formulate a unified standard of establishing reference ESR values, this paper presents a novel prediction model in which local normal ESR values and corresponding geographical factors are used to predict reference ESR values using multi-layer feed-forward artificial neural networks (ANN). Methods and findings Local normal ESR values were obtained from hospital data, while geographical factors that include altitude, sunshine hours, relative humidity, temperature and precipitation were obtained from the National Geographical Data Information Centre in China. The results show that predicted values are statistically in agreement with measured values. Model results exhibit significant agreement between training data and test data. Consequently, the model is used to predict the unseen local reference ESR values. Conclusions Reference ESR values can be established with geographical factors by using artificial intelligence techniques. ANN is an effective method for simulating and predicting reference ESR values because of its ability to model nonlinear and complex relationships. PMID:23497145

Incorporating geographical factors with artificial neural networks to predict reference values of erythrocyte sedimentation rate.

PubMed

Yang, Qingsheng; Mwenda, Kevin M; Ge, Miao

2013-03-12

The measurement of the Erythrocyte Sedimentation Rate (ESR) value is a standard procedure performed during a typical blood test. In order to formulate a unified standard of establishing reference ESR values, this paper presents a novel prediction model in which local normal ESR values and corresponding geographical factors are used to predict reference ESR values using multi-layer feed-forward artificial neural networks (ANN). Local normal ESR values were obtained from hospital data, while geographical factors that include altitude, sunshine hours, relative humidity, temperature and precipitation were obtained from the National Geographical Data Information Centre in China.The results show that predicted values are statistically in agreement with measured values. Model results exhibit significant agreement between training data and test data. Consequently, the model is used to predict the unseen local reference ESR values. Reference ESR values can be established with geographical factors by using artificial intelligence techniques. ANN is an effective method for simulating and predicting reference ESR values because of its ability to model nonlinear and complex relationships.

A Hybrid Neuro-Fuzzy Model For Integrating Large Earth-Science Datasets

NASA Astrophysics Data System (ADS)

Porwal, A.; Carranza, J.; Hale, M.

2004-12-01

A GIS-based hybrid neuro-fuzzy approach to integration of large earth-science datasets for mineral prospectivity mapping is described. It implements a Takagi-Sugeno type fuzzy inference system in the framework of a four-layered feed-forward adaptive neural network. Each unique combination of the datasets is considered a feature vector whose components are derived by knowledge-based ordinal encoding of the constituent datasets. A subset of feature vectors with a known output target vector (i.e., unique conditions known to be associated with either a mineralized or a barren location) is used for the training of an adaptive neuro-fuzzy inference system. Training involves iterative adjustment of parameters of the adaptive neuro-fuzzy inference system using a hybrid learning procedure for mapping each training vector to its output target vector with minimum sum of squared error. The trained adaptive neuro-fuzzy inference system is used to process all feature vectors. The output for each feature vector is a value that indicates the extent to which a feature vector belongs to the mineralized class or the barren class. These values are used to generate a prospectivity map. The procedure is demonstrated by an application to regional-scale base metal prospectivity mapping in a study area located in the Aravalli metallogenic province (western India). A comparison of the hybrid neuro-fuzzy approach with pure knowledge-driven fuzzy and pure data-driven neural network approaches indicates that the former offers a superior method for integrating large earth-science datasets for predictive spatial mathematical modelling.

Feed-Forward Neural Network Soft-Sensor Modeling of Flotation Process Based on Particle Swarm Optimization and Gravitational Search Algorithm

PubMed Central

Wang, Jie-Sheng; Han, Shuang

2015-01-01

For predicting the key technology indicators (concentrate grade and tailings recovery rate) of flotation process, a feed-forward neural network (FNN) based soft-sensor model optimized by the hybrid algorithm combining particle swarm optimization (PSO) algorithm and gravitational search algorithm (GSA) is proposed. Although GSA has better optimization capability, it has slow convergence velocity and is easy to fall into local optimum. So in this paper, the velocity vector and position vector of GSA are adjusted by PSO algorithm in order to improve its convergence speed and prediction accuracy. Finally, the proposed hybrid algorithm is adopted to optimize the parameters of FNN soft-sensor model. Simulation results show that the model has better generalization and prediction accuracy for the concentrate grade and tailings recovery rate to meet the online soft-sensor requirements of the real-time control in the flotation process. PMID:26583034

Optimal Decision Making in Neural Inhibition Models

ERIC Educational Resources Information Center

van Ravenzwaaij, Don; van der Maas, Han L. J.; Wagenmakers, Eric-Jan

2012-01-01

In their influential "Psychological Review" article, Bogacz, Brown, Moehlis, Holmes, and Cohen (2006) discussed optimal decision making as accomplished by the drift diffusion model (DDM). The authors showed that neural inhibition models, such as the leaky competing accumulator model (LCA) and the feedforward inhibition model (FFI), can mimic the…

Decoupling feedforward and feedback structures in hybrid active noise control systems for uncorrelated narrowband disturbances

NASA Astrophysics Data System (ADS)

Wu, Lifu; Qiu, Xiaojun; Burnett, Ian S.; Guo, Yecai

2015-08-01

Hybrid feedforward and feedback structures are useful for active noise control (ANC) applications where the noise can only be partially obtained with reference sensors. The traditional method uses the secondary signals of both the feedforward and feedback structures to synthesize a reference signal for the feedback structure in the hybrid structure. However, this approach introduces coupling between the feedforward and feedback structures and parameter changes in one structure affect the other during adaptation such that the feedforward and feedback structures must be optimized simultaneously in practical ANC system design. Two methods are investigated in this paper to remove such coupling effects. One is a simplified method, which uses the error signal directly as the reference signal in the feedback structure, and the second method generates the reference signal for the feedback structure by using only the secondary signal from the feedback structure and utilizes the generated reference signal as the error signal of the feedforward structure. Because the two decoupling methods can optimize the feedforward and feedback structures separately, they provide more flexibility in the design and optimization of the adaptive filters in practical ANC applications.

Estimation of Thalamocortical and Intracortical Network Models from Joint Thalamic Single-Electrode and Cortical Laminar-Electrode Recordings in the Rat Barrel System

PubMed Central

Blomquist, Patrick; Devor, Anna; Indahl, Ulf G.; Ulbert, Istvan; Einevoll, Gaute T.; Dale, Anders M.

2009-01-01

A new method is presented for extraction of population firing-rate models for both thalamocortical and intracortical signal transfer based on stimulus-evoked data from simultaneous thalamic single-electrode and cortical recordings using linear (laminar) multielectrodes in the rat barrel system. Time-dependent population firing rates for granular (layer 4), supragranular (layer 2/3), and infragranular (layer 5) populations in a barrel column and the thalamic population in the homologous barreloid are extracted from the high-frequency portion (multi-unit activity; MUA) of the recorded extracellular signals. These extracted firing rates are in turn used to identify population firing-rate models formulated as integral equations with exponentially decaying coupling kernels, allowing for straightforward transformation to the more common firing-rate formulation in terms of differential equations. Optimal model structures and model parameters are identified by minimizing the deviation between model firing rates and the experimentally extracted population firing rates. For the thalamocortical transfer, the experimental data favor a model with fast feedforward excitation from thalamus to the layer-4 laminar population combined with a slower inhibitory process due to feedforward and/or recurrent connections and mixed linear-parabolic activation functions. The extracted firing rates of the various cortical laminar populations are found to exhibit strong temporal correlations for the present experimental paradigm, and simple feedforward population firing-rate models combined with linear or mixed linear-parabolic activation function are found to provide excellent fits to the data. The identified thalamocortical and intracortical network models are thus found to be qualitatively very different. While the thalamocortical circuit is optimally stimulated by rapid changes in the thalamic firing rate, the intracortical circuits are low-pass and respond most strongly to slowly varying inputs from the cortical layer-4 population. PMID:19325875

Context-based virtual metrology

NASA Astrophysics Data System (ADS)

Ebersbach, Peter; Urbanowicz, Adam M.; Likhachev, Dmitriy; Hartig, Carsten; Shifrin, Michael

2018-03-01

Hybrid and data feed forward methodologies are well established for advanced optical process control solutions in highvolume semiconductor manufacturing. Appropriate information from previous measurements, transferred into advanced optical model(s) at following step(s), provides enhanced accuracy and exactness of the measured topographic (thicknesses, critical dimensions, etc.) and material parameters. In some cases, hybrid or feed-forward data are missed or invalid for dies or for a whole wafer. We focus on approaches of virtual metrology to re-create hybrid or feed-forward data inputs in high-volume manufacturing. We discuss missing data inputs reconstruction which is based on various interpolation and extrapolation schemes and uses information about wafer's process history. Moreover, we demonstrate data reconstruction approach based on machine learning techniques utilizing optical model and measured spectra. And finally, we investigate metrics that allow one to assess error margin of virtual data input.

Measuring Feedforward Inhibition and Its Impact on Local Circuit Function.

PubMed

Hull, Court

2017-05-01

This protocol describes a series of approaches to measure feedforward inhibition in acute brain slices from the cerebellar cortex. Using whole-cell voltage and current clamp recordings from Purkinje cells in conjunction with electrical stimulation of the parallel fibers, these methods demonstrate how to measure the relationship between excitation and inhibition in a feedforward circuit. This protocol also describes how to measure the impact of feedforward inhibition on Purkinje cell excitability, with an emphasis on spike timing. © 2017 Cold Spring Harbor Laboratory Press.

Overreliance on auditory feedback may lead to sound/syllable repetitions: simulations of stuttering and fluency-inducing conditions with a neural model of speech production.

PubMed

Civier, Oren; Tasko, Stephen M; Guenther, Frank H

2010-09-01

This paper investigates the hypothesis that stuttering may result in part from impaired readout of feedforward control of speech, which forces persons who stutter (PWS) to produce speech with a motor strategy that is weighted too much toward auditory feedback control. Over-reliance on feedback control leads to production errors which if they grow large enough, can cause the motor system to "reset" and repeat the current syllable. This hypothesis is investigated using computer simulations of a "neurally impaired" version of the DIVA model, a neural network model of speech acquisition and production. The model's outputs are compared to published acoustic data from PWS' fluent speech, and to combined acoustic and articulatory movement data collected from the dysfluent speech of one PWS. The simulations mimic the errors observed in the PWS subject's speech, as well as the repairs of these errors. Additional simulations were able to account for enhancements of fluency gained by slowed/prolonged speech and masking noise. Together these results support the hypothesis that many dysfluencies in stuttering are due to a bias away from feedforward control and toward feedback control. The reader will be able to (a) describe the contribution of auditory feedback control and feedforward control to normal and stuttered speech production, (b) summarize the neural modeling approach to speech production and its application to stuttering, and (c) explain how the DIVA model accounts for enhancements of fluency gained by slowed/prolonged speech and masking noise.

A hybrid prognostic model for multistep ahead prediction of machine condition

NASA Astrophysics Data System (ADS)

Roulias, D.; Loutas, T. H.; Kostopoulos, V.

2012-05-01

Prognostics are the future trend in condition based maintenance. In the current framework a data driven prognostic model is developed. The typical procedure of developing such a model comprises a) the selection of features which correlate well with the gradual degradation of the machine and b) the training of a mathematical tool. In this work the data are taken from a laboratory scale single stage gearbox under multi-sensor monitoring. Tests monitoring the condition of the gear pair from healthy state until total brake down following several days of continuous operation were conducted. After basic pre-processing of the derived data, an indicator that correlated well with the gearbox condition was obtained. Consecutively the time series is split in few distinguishable time regions via an intelligent data clustering scheme. Each operating region is modelled with a feed-forward artificial neural network (FFANN) scheme. The performance of the proposed model is tested by applying the system to predict the machine degradation level on unseen data. The results show the plausibility and effectiveness of the model in following the trend of the timeseries even in the case that a sudden change occurs. Moreover the model shows ability to generalise for application in similar mechanical assets.

Glucocorticoid and cytokine crosstalk: Feedback, feedforward, and co-regulatory interactions determine repression or resistance

PubMed Central

Shah, Suharsh; Altonsy, Mohammed O.; Gerber, Antony N.

2017-01-01

Inflammatory signals induce feedback and feedforward systems that provide temporal control. Although glucocorticoids can repress inflammatory gene expression, glucocorticoid receptor recruitment increases expression of negative feedback and feedforward regulators, including the phosphatase, DUSP1, the ubiquitin-modifying enzyme, TNFAIP3, or the mRNA-destabilizing protein, ZFP36. Moreover, glucocorticoid receptor cooperativity with factors, including nuclear factor-κB (NF-κB), may enhance regulator expression to promote repression. Conversely, MAPKs, which are inhibited by glucocorticoids, provide feedforward control to limit expression of the transcription factor IRF1, and the chemokine, CXCL10. We propose that modulation of feedback and feedforward control can determine repression or resistance of inflammatory gene expression toglucocorticoid. PMID:28283576

Min-max hyperellipsoidal clustering for anomaly detection in network security.

PubMed

Sarasamma, Suseela T; Zhu, Qiuming A

2006-08-01

A novel hyperellipsoidal clustering technique is presented for an intrusion-detection system in network security. Hyperellipsoidal clusters toward maximum intracluster similarity and minimum intercluster similarity are generated from training data sets. The novelty of the technique lies in the fact that the parameters needed to construct higher order data models in general multivariate Gaussian functions are incrementally derived from the data sets using accretive processes. The technique is implemented in a feedforward neural network that uses a Gaussian radial basis function as the model generator. An evaluation based on the inclusiveness and exclusiveness of samples with respect to specific criteria is applied to accretively learn the output clusters of the neural network. One significant advantage of this is its ability to detect individual anomaly types that are hard to detect with other anomaly-detection schemes. Applying this technique, several feature subsets of the tcptrace network-connection records that give above 95% detection at false-positive rates below 5% were identified.

Prediction of municipal solid waste generation using nonlinear autoregressive network.

PubMed

Younes, Mohammad K; Nopiah, Z M; Basri, N E Ahmad; Basri, H; Abushammala, Mohammed F M; Maulud, K N A

2015-12-01

Most of the developing countries have solid waste management problems. Solid waste strategic planning requires accurate prediction of the quality and quantity of the generated waste. In developing countries, such as Malaysia, the solid waste generation rate is increasing rapidly, due to population growth and new consumption trends that characterize society. This paper proposes an artificial neural network (ANN) approach using feedforward nonlinear autoregressive network with exogenous inputs (NARX) to predict annual solid waste generation in relation to demographic and economic variables like population number, gross domestic product, electricity demand per capita and employment and unemployment numbers. In addition, variable selection procedures are also developed to select a significant explanatory variable. The model evaluation was performed using coefficient of determination (R(2)) and mean square error (MSE). The optimum model that produced the lowest testing MSE (2.46) and the highest R(2) (0.97) had three inputs (gross domestic product, population and employment), eight neurons and one lag in the hidden layer, and used Fletcher-Powell's conjugate gradient as the training algorithm.

Virtual geotechnical laboratory experiments using a simulator

NASA Astrophysics Data System (ADS)

Penumadu, Dayakar; Zhao, Rongda; Frost, David

2000-04-01

The details of a test simulator that provides a realistic environment for performing virtual laboratory experimentals in soil mechanics is presented. A computer program Geo-Sim that can be used to perform virtual experiments, and allow for real-time observations of material response is presented. The results of experiments, for a given set of input parameters, are obtained with the test simulator using well-trained artificial neural-network-based soil models for different soil types and stress paths. Multimedia capabilities are integrated in Geo-Sim, using software that links and controls a laser disc player with a real-time parallel processing ability. During the simulation of a virtual experiment, relevant portions of the video image of a previously recorded test on an actual soil specimen are dispalyed along with the graphical presentation of response from the feedforward ANN model predictions. The pilot simulator developed to date includes all aspects related to performing a triaxial test on cohesionless soil under undrained and drained conditions. The benefits of the test simulator are also presented.

Feedforward Controller of Ill-Conditioned Hysteresis Using Singularity-Free Prandtl–Ishlinskii Model

PubMed Central

Tan, U-Xuan; Latt, Win Tun; Shee, Cheng Yap; Riviere, Cameron N.; Ang, Wei Tech

2009-01-01

Piezoelectric, magnetostrictive, and shape memory alloy actuators are gaining importance in high-frequency precision applications constrained by space. Their intrinsic hysteretic behavior makes control difficult. The Prandtl–Ishlinskii (PI) operator can model hysteresis well, albeit a major inadequacy: the inverse operator does not exist when the hysteretic curve gradient is not positive definite, i.e., ill condition occurs when slope is negative. An inevitable tradeoff between modeling accuracy and inversion stability exists. The hysteretic modeling improves with increasing number of play operators. But as the piecewise continuous interval of each operator reduces, the model tends to be ill-conditioned, especially at the turning points. Similar ill-conditioned situation arises when these actuators move heavy loads or operate at high frequency. This paper proposes an extended PI operator to map hysteresis to a domain where inversion is well behaved. The inverse weights are then evaluated to determine the inverse hysteresis model for the feedforward controller. For illustration purpose, a piezoelectric actuator is used. PMID:19936032

Feedforward Inhibition Allows Input Summation to Vary in Recurrent Cortical Networks

PubMed Central

2018-01-01

Abstract Brain computations depend on how neurons transform inputs to spike outputs. Here, to understand input-output transformations in cortical networks, we recorded spiking responses from visual cortex (V1) of awake mice of either sex while pairing sensory stimuli with optogenetic perturbation of excitatory and parvalbumin-positive inhibitory neurons. We found that V1 neurons’ average responses were primarily additive (linear). We used a recurrent cortical network model to determine whether these data, as well as past observations of nonlinearity, could be described by a common circuit architecture. Simulations showed that cortical input-output transformations can be changed from linear to sublinear with moderate (∼20%) strengthening of connections between inhibitory neurons, but this change away from linear scaling depends on the presence of feedforward inhibition. Simulating a variety of recurrent connection strengths showed that, compared with when input arrives only to excitatory neurons, networks produce a wider range of output spiking responses in the presence of feedforward inhibition. PMID:29682603

Practical training framework for fitting a function and its derivatives.

PubMed

Pukrittayakamee, Arjpolson; Hagan, Martin; Raff, Lionel; Bukkapatnam, Satish T S; Komanduri, Ranga

2011-06-01

This paper describes a practical framework for using multilayer feedforward neural networks to simultaneously fit both a function and its first derivatives. This framework involves two steps. The first step is to train the network to optimize a performance index, which includes both the error in fitting the function and the error in fitting the derivatives. The second step is to prune the network by removing neurons that cause overfitting and then to retrain it. This paper describes two novel types of overfitting that are only observed when simultaneously fitting both a function and its first derivatives. A new pruning algorithm is proposed to eliminate these types of overfitting. Experimental results show that the pruning algorithm successfully eliminates the overfitting and produces the smoothest responses and the best generalization among all the training algorithms that we have tested.

Functional electrical stimulation controlled by artificial neural networks: pilot experiments with simple movements are promising for rehabilitation applications.

PubMed

Ferrante, Simona; Pedrocchi, Alessandra; Iannò, Marco; De Momi, Elena; Ferrarin, Maurizio; Ferrigno, Giancarlo

2004-01-01

This study falls within the ambit of research on functional electrical stimulation for the design of rehabilitation training for spinal cord injured patients. In this context, a crucial issue is the control of the stimulation parameters in order to optimize the patterns of muscle activation and to increase the duration of the exercises. An adaptive control system (NEURADAPT) based on artificial neural networks (ANNs) was developed to control the knee joint in accordance with desired trajectories by stimulating quadriceps muscles. This strategy includes an inverse neural model of the stimulated limb in the feedforward line and a neural network trained on-line in the feedback loop. NEURADAPT was compared with a linear closed-loop proportional integrative derivative (PID) controller and with a model-based neural controller (NEUROPID). Experiments on two subjects (one healthy and one paraplegic) show the good performance of NEURADAPT, which is able to reduce the time lag introduced by the PID controller. In addition, control systems based on ANN techniques do not require complicated calibration procedures at the beginning of each experimental session. After the initial learning phase, the ANN, thanks to its generalization capacity, is able to cope with a certain range of variability of skeletal muscle properties.

Stabilization of burn conditions in a thermonuclear reactor using artificial neural networks

NASA Astrophysics Data System (ADS)

Vitela, Javier E.; Martinell, Julio J.

1998-02-01

In this work we develop an artificial neural network (ANN) for the feedback stabilization of a thermonuclear reactor at nearly ignited burn conditions. A volume-averaged zero-dimensional nonlinear model is used to represent the time evolution of the electron density, the relative density of alpha particles and the temperature of the plasma, where a particular scaling law for the energy confinement time previously used by other authors, was adopted. The control actions include the concurrent modulation of the D-T refuelling rate, the injection of a neutral He-4 beam and an auxiliary heating power modulation, which are constrained to take values within a maximum and minimum levels. For this purpose a feedforward multilayer artificial neural network with sigmoidal activation function is trained using a back-propagation through-time technique. Numerical examples are used to illustrate the behaviour of the resulting ANN-dynamical system configuration. It is concluded that the resulting ANN can successfully stabilize the nonlinear model of the thermonuclear reactor at nearly ignited conditions for temperature and density departures significantly far from their nominal operating values. The NN-dynamical system configuration is shown to be robust with respect to the thermalization time of the alpha particles for perturbations within the region used to train the NN.

Recurrent Convolutional Neural Networks: A Better Model of Biological Object Recognition.

PubMed

Spoerer, Courtney J; McClure, Patrick; Kriegeskorte, Nikolaus

2017-01-01

Feedforward neural networks provide the dominant model of how the brain performs visual object recognition. However, these networks lack the lateral and feedback connections, and the resulting recurrent neuronal dynamics, of the ventral visual pathway in the human and non-human primate brain. Here we investigate recurrent convolutional neural networks with bottom-up (B), lateral (L), and top-down (T) connections. Combining these types of connections yields four architectures (B, BT, BL, and BLT), which we systematically test and compare. We hypothesized that recurrent dynamics might improve recognition performance in the challenging scenario of partial occlusion. We introduce two novel occluded object recognition tasks to test the efficacy of the models, digit clutter (where multiple target digits occlude one another) and digit debris (where target digits are occluded by digit fragments). We find that recurrent neural networks outperform feedforward control models (approximately matched in parametric complexity) at recognizing objects, both in the absence of occlusion and in all occlusion conditions. Recurrent networks were also found to be more robust to the inclusion of additive Gaussian noise. Recurrent neural networks are better in two respects: (1) they are more neurobiologically realistic than their feedforward counterparts; (2) they are better in terms of their ability to recognize objects, especially under challenging conditions. This work shows that computer vision can benefit from using recurrent convolutional architectures and suggests that the ubiquitous recurrent connections in biological brains are essential for task performance.

Asymptotic tracking and disturbance rejection of the blood glucose regulation system.

PubMed

Ashley, Brandon; Liu, Weijiu

2017-07-01

Type 1 diabetes patients need external insulin to maintain blood glucose within a narrow range from 65 to 108 mg/dl (3.6 to 6.0 mmol/l). A mathematical model for the blood glucose regulation is required for integrating a glucose monitoring system into insulin pump technology to form a closed-loop insulin delivery system on the feedback of the blood glucose, the so-called "artificial pancreas". The objective of this paper is to treat the exogenous glucose from food as a glucose disturbance and then develop a closed-loop feedback and feedforward control system for the blood glucose regulation system subject to the exogenous glucose disturbance. For this, a mathematical model for the glucose disturbance is proposed on the basis of experimental data, and then incorporated into an existing blood glucose regulation model. Because all the eigenvalues of the disturbance model have zero real parts, the center manifold theory is used to establish blood glucose regulator equations. We then use their solutions to synthesize a required feedback and feedforward controller to reject the disturbance and asymptotically track a constant glucose reference of 90  mg/dl. Since the regulator equations are nonlinear partial differential equations and usually impossible to solve analytically, a linear approximation solution is obtained. Our numerical simulations show that, under the linear approximate feedback and feedforward controller, the blood glucose asymptotically tracks its desired level of 90 mg/dl approximately. Copyright © 2017 Elsevier Inc. All rights reserved.

Application of a feedforward neural network in the search for kuroko deposits in the Hokuroku District, Japan

USGS Publications Warehouse

Singer, Donald A.; Kouda, Ryoichi

1996-01-01

A feedforward neural network with one hidden layer and five neurons was trained to recognize the distance to kuroko mineral deposits. Average amounts per hole of pyrite, sericite, and gypsum plus anhydrite as measured by X-rays in 69 drillholes were used to train the net. Drillholes near and between the Fukazawa, Furutobe, and Shakanai mines were used. The training data were selected carefully to represent well-explored areas where some confidence of the distance to ore was assured. A logarithmic transform was applied to remove the skewness of distance and each variable was scaled and centered by subtracting the median and dividing by the interquartile range. The learning algorithm of annealing plus conjugate gradients was used to minimize the mean squared error of the scaled distance to ore. The trained network then was applied to all of the 152 drillholes that had measured gypsum, sericite, and pyrite. A contour plot of the neural net predicted distance to ore shows fairly wide areas of 1 km or less to ore; each of the known deposit groups is within the 1 km contour. The high and low distances on the margins of the contoured distance plot are in part the result of boundary effects of the contouring algorithm. For example, the short distances to ore predicted west of the Shakanai (Hanaoka) deposits are in basement. However, the short distances to ore predicted northeast of Furotobe, just off the figure, coincide with the location of the Nurukawa kuroko deposit and the Omaki deposit, south of the Shakanai-Hanaoka deposits, seems to be on an extension of short distance to ore contour, but is beyond the 3 km limit from drillholes. Also of interest are some areas only a few kilometers from the Fukazawa and Shakanai groups of deposits that are estimated to be many kilometers from ore, apparently reflecting the network's recognition of the extreme local variability of the geology near some deposits.

Groundwater-level prediction using multiple linear regression and artificial neural network techniques: a comparative assessment

NASA Astrophysics Data System (ADS)

Sahoo, Sasmita; Jha, Madan K.

2013-12-01

The potential of multiple linear regression (MLR) and artificial neural network (ANN) techniques in predicting transient water levels over a groundwater basin were compared. MLR and ANN modeling was carried out at 17 sites in Japan, considering all significant inputs: rainfall, ambient temperature, river stage, 11 seasonal dummy variables, and influential lags of rainfall, ambient temperature, river stage and groundwater level. Seventeen site-specific ANN models were developed, using multi-layer feed-forward neural networks trained with Levenberg-Marquardt backpropagation algorithms. The performance of the models was evaluated using statistical and graphical indicators. Comparison of the goodness-of-fit statistics of the MLR models with those of the ANN models indicated that there is better agreement between the ANN-predicted groundwater levels and the observed groundwater levels at all the sites, compared to the MLR. This finding was supported by the graphical indicators and the residual analysis. Thus, it is concluded that the ANN technique is superior to the MLR technique in predicting spatio-temporal distribution of groundwater levels in a basin. However, considering the practical advantages of the MLR technique, it is recommended as an alternative and cost-effective groundwater modeling tool.

Feedforward Coordinate Control of a Robotic Cell Injection Catheter.

PubMed

Cheng, Weyland; Law, Peter K

2017-08-01

Remote and robotically actuated catheters are the stepping-stones toward autonomous catheters, where complex intravascular procedures may be performed with minimal intervention from a physician. This article proposes a concept for the positional, feedforward control of a robotically actuated cell injection catheter used for the injection of myogenic or undifferentiated stem cells into the myocardial infarct boundary zones of the left ventricle. The prototype for the catheter system was built upon a needle-based catheter with a single degree of deflection, a 3-D printed handle combined with actuators, and the Arduino microcontroller platform. A bench setup was used to mimic a left ventricle catheter procedure starting from the femoral artery. Using Matlab and the open-source video modeling tool Tracker, the planar coordinates ( y, z) of the catheter position were analyzed, and a feedforward control system was developed based on empirical models. Using the Student's t test with a sample size of 26, it was determined that for both the y- and z-axes, the mean discrepancy between the calibrated and theoretical coordinate values had no significant difference compared to the hypothetical value of µ = 0. The root mean square error of the calibrated coordinates also showed an 88% improvement in the z-axis and 31% improvement in the y-axis compared to the unmodified trial run. This proof of concept investigation leads to the possibility of further developing a feedfoward control system in vivo using catheters with omnidirectional deflection. Feedforward positional control allows for more flexibility in the design of an automated catheter system where problems such as systemic time delay may be a hindrance in instances requiring an immediate reaction.

Adaptive control for accelerators

DOEpatents

Eaton, Lawrie E.; Jachim, Stephen P.; Natter, Eckard F.

1991-01-01

An adaptive feedforward control loop is provided to stabilize accelerator beam loading of the radio frequency field in an accelerator cavity during successive pulses of the beam into the cavity. A digital signal processor enables an adaptive algorithm to generate a feedforward error correcting signal functionally determined by the feedback error obtained by a beam pulse loading the cavity after the previous correcting signal was applied to the cavity. Each cavity feedforward correcting signal is successively stored in the digital processor and modified by the feedback error resulting from its application to generate the next feedforward error correcting signal. A feedforward error correcting signal is generated by the digital processor in advance of the beam pulse to enable a composite correcting signal and the beam pulse to arrive concurrently at the cavity.

Glucocorticoid and cytokine crosstalk: Feedback, feedforward, and co-regulatory interactions determine repression or resistance.

PubMed

Newton, Robert; Shah, Suharsh; Altonsy, Mohammed O; Gerber, Antony N

2017-04-28

Inflammatory signals induce feedback and feedforward systems that provide temporal control. Although glucocorticoids can repress inflammatory gene expression, glucocorticoid receptor recruitment increases expression of negative feedback and feedforward regulators, including the phosphatase, DUSP1, the ubiquitin-modifying enzyme, TNFAIP3, or the mRNA-destabilizing protein, ZFP36. Moreover, glucocorticoid receptor cooperativity with factors, including nuclear factor-κB (NF-κB), may enhance regulator expression to promote repression. Conversely, MAPKs, which are inhibited by glucocorticoids, provide feedforward control to limit expression of the transcription factor IRF1, and the chemokine, CXCL10. We propose that modulation of feedback and feedforward control can determine repression or resistance of inflammatory gene expression toglucocorticoid. © 2017 by The American Society for Biochemistry and Molecular Biology, Inc.

Feedforward control of a closed-loop piezoelectric translation stage for atomic force microscope.

PubMed

Li, Yang; Bechhoefer, John

2007-01-01

Simple feedforward ideas are shown to lead to a nearly tenfold increase in the effective bandwidth of a closed-loop piezoelectric positioning stage used in scanning probe microscopy. If the desired control signal is known in advance, the feedforward filter can be acausal: the information about the future can be used to make the output of the stage have almost no phase lag with respect to the input. This keeps in register the images assembled from right and left scans. We discuss the design constraints imposed by the need for the feedforward filter to work robustly under a variety of circumstances. Because the feedforward needs only to modify the input signal, it can be added to any piezoelectric stage, whether closed or open loop.

Research on Modeling of the Agile Satellite Using a Single Gimbal Magnetically Suspended CMG and the Disturbance Feedforward Compensation for Rotors

PubMed Central

Cui, Peiling; Yan, Ning

2012-01-01

The magnetically suspended Control Moment Gyroscope (CMG) has the advantages of long-life, micro-vibration and being non-lubricating, and is the ideal actuator for agile maneuver satellite attitude control. However, the stability of the rotor in magnetic bearing and the precision of the output torque of a magnetically suspended CMG are affected by the rapid maneuvers of satellites. In this paper, a dynamic model of the agile satellite including a magnetically suspended single gimbal control moment gyroscope is built and the equivalent disturbance torque effected on the rotor is obtained. The feedforward compensation control method is used to depress the disturbance on the rotor. Simulation results are given to show that the rotor displacement is obviously reduced. PMID:23235442

A novel memristive multilayer feedforward small-world neural network with its applications in PID control.

PubMed

Dong, Zhekang; Duan, Shukai; Hu, Xiaofang; Wang, Lidan; Li, Hai

2014-01-01

In this paper, we present an implementation scheme of memristor-based multilayer feedforward small-world neural network (MFSNN) inspirited by the lack of the hardware realization of the MFSNN on account of the need of a large number of electronic neurons and synapses. More specially, a mathematical closed-form charge-governed memristor model is presented with derivation procedures and the corresponding Simulink model is presented, which is an essential block for realizing the memristive synapse and the activation function in electronic neurons. Furthermore, we investigate a more intelligent memristive PID controller by incorporating the proposed MFSNN into intelligent PID control based on the advantages of the memristive MFSNN on computation speed and accuracy. Finally, numerical simulations have demonstrated the effectiveness of the proposed scheme.

PIFCGT: A PIF autopilot design program for general aviation aircraft

NASA Technical Reports Server (NTRS)

Broussard, J. R.

1983-01-01

This report documents the PIFCGT computer program. In FORTRAN, PIFCGT is a computer design aid for determing Proportional-Integral-Filter (PIF) control laws for aircraft autopilots implemented with a Command Generator Tracker (CGT). The program uses Linear-Quadratic-Regulator synthesis algorithms to determine feedback gains, and includes software to solve the feedforward matrix equation which is useful in determining the command generator tracker feedforward gains. The program accepts aerodynamic stability derivatives and computes the corresponding aerodynamic linear model. The nine autopilot modes that can be designed include four maneuver modes (ROLL SEL, PITCH SEL, HDG SEL, ALT SEL), four final approach models (APR GS, APR LOCI, APR LOCR, APR LOCP), and a BETA HOLD mode. The program has been compiled and executed on a CDC computer.

A Novel Memristive Multilayer Feedforward Small-World Neural Network with Its Applications in PID Control

PubMed Central

Dong, Zhekang; Duan, Shukai; Hu, Xiaofang; Wang, Lidan

2014-01-01

In this paper, we present an implementation scheme of memristor-based multilayer feedforward small-world neural network (MFSNN) inspirited by the lack of the hardware realization of the MFSNN on account of the need of a large number of electronic neurons and synapses. More specially, a mathematical closed-form charge-governed memristor model is presented with derivation procedures and the corresponding Simulink model is presented, which is an essential block for realizing the memristive synapse and the activation function in electronic neurons. Furthermore, we investigate a more intelligent memristive PID controller by incorporating the proposed MFSNN into intelligent PID control based on the advantages of the memristive MFSNN on computation speed and accuracy. Finally, numerical simulations have demonstrated the effectiveness of the proposed scheme. PMID:25202723

A unified perspective on robot control - The energy Lyapunov function approach

NASA Technical Reports Server (NTRS)

Wen, John T.

1990-01-01

A unified framework for the stability analysis of robot tracking control is presented. By using an energy-motivated Lyapunov function candidate, the closed-loop stability is shown for a large family of control laws sharing a common structure of proportional and derivative feedback and a model-based feedforward. The feedforward can be zero, partial or complete linearized dynamics, partial or complete nonlinear dynamics, or linearized or nonlinear dynamics with parameter adaptation. As result, the dichotomous approaches to the robot control problem based on the open-loop linearization and nonlinear Lyapunov analysis are both included in this treatment. Furthermore, quantitative estimates of the trade-offs between different schemes in terms of the tracking performance, steady state error, domain of convergence, realtime computation load and required a prior model information are derived.

Research on modeling of the agile satellite using a single gimbal magnetically suspended CMG and the disturbance feedforward compensation for rotors.

PubMed

Cui, Peiling; Yan, Ning

2012-12-12

The magnetically suspended Control Moment Gyroscope (CMG) has the advantages of long-life, micro-vibration and being non-lubricating, and is the ideal actuator for agile maneuver satellite attitude control. However, the stability of the rotor in magnetic bearing and the precision of the output torque of a magnetically suspended CMG are affected by the rapid maneuvers of satellites. In this paper, a dynamic model of the agile satellite including a magnetically suspended single gimbal control moment gyroscope is built and the equivalent disturbance torque effected on the rotor is obtained. The feedforward compensation control method is used to depress the disturbance on the rotor. Simulation results are given to show that the rotor displacement is obviously reduced.

Estimating feedforward vs. feedback control of speech production through kinematic analyses of unperturbed articulatory movements.

PubMed

Kim, Kwang S; Max, Ludo

2014-01-01

To estimate the contributions of feedforward vs. feedback control systems in speech articulation, we analyzed the correspondence between initial and final kinematics in unperturbed tongue and jaw movements for consonant-vowel (CV) and vowel-consonant (VC) syllables. If movement extents and endpoints are highly predictable from early kinematic information, then the movements were most likely completed without substantial online corrections (feedforward control); if the correspondence between early kinematics and final amplitude or position is low, online adjustments may have altered the planned trajectory (feedback control) (Messier and Kalaska, 1999). Five adult speakers produced CV and VC syllables with high, mid, or low vowels while movements of the tongue and jaw were tracked electromagnetically. The correspondence between the kinematic parameters peak acceleration or peak velocity and movement extent as well as between the articulators' spatial coordinates at those kinematic landmarks and movement endpoint was examined both for movements across different target distances (i.e., across vowel height) and within target distances (i.e., within vowel height). Taken together, results suggest that jaw and tongue movements for these CV and VC syllables are mostly under feedforward control but with feedback-based contributions. One type of feedback-driven compensatory adjustment appears to regulate movement duration based on variation in peak acceleration. Results from a statistical model based on multiple regression are presented to illustrate how the relative strength of these feedback contributions can be estimated.

Inhibitory Gating of Basolateral Amygdala Inputs to the Prefrontal Cortex

PubMed Central

McGarry, Laura M.

2016-01-01

Interactions between the prefrontal cortex (PFC) and basolateral amygdala (BLA) regulate emotional behaviors. However, a circuit-level understanding of functional connections between these brain regions remains incomplete. The BLA sends prominent glutamatergic projections to the PFC, but the overall influence of these inputs is predominantly inhibitory. Here we combine targeted recordings and optogenetics to examine the synaptic underpinnings of this inhibition in the mouse infralimbic PFC. We find that BLA inputs preferentially target layer 2 corticoamygdala over neighboring corticostriatal neurons. However, these inputs make even stronger connections onto neighboring parvalbumin and somatostatin expressing interneurons. Inhibitory connections from these two populations of interneurons are also much stronger onto corticoamygdala neurons. Consequently, BLA inputs are able to drive robust feedforward inhibition via two parallel interneuron pathways. Moreover, the contributions of these interneurons shift during repetitive activity, due to differences in short-term synaptic dynamics. Thus, parvalbumin interneurons are activated at the start of stimulus trains, whereas somatostatin interneuron activation builds during these trains. Together, these results reveal how the BLA impacts the PFC through a complex interplay of direct excitation and feedforward inhibition. They also highlight the roles of targeted connections onto multiple projection neurons and interneurons in this cortical circuit. Our findings provide a mechanistic understanding for how the BLA can influence the PFC circuit, with important implications for how this circuit participates in the regulation of emotion. SIGNIFICANCE STATEMENT The prefrontal cortex (PFC) and basolateral amygdala (BLA) interact to control emotional behaviors. Here we show that BLA inputs elicit direct excitation and feedforward inhibition of layer 2 projection neurons in infralimbic PFC. BLA inputs are much stronger at corticoamygdala neurons compared with nearby corticostriatal neurons. However, these inputs are even more powerful at parvalbumin and somatostatin expressing interneurons. BLA inputs thus activate two parallel inhibitory networks, whose contributions change during repetitive activity. Finally, connections from these interneurons are also more powerful at corticoamygdala neurons compared with corticostriatal neurons. Together, our results demonstrate how the BLA predominantly inhibits the PFC via a complex sequence involving multiple cell-type and input-specific connections. PMID:27605614

Inhibitory Gating of Basolateral Amygdala Inputs to the Prefrontal Cortex.

PubMed

McGarry, Laura M; Carter, Adam G

2016-09-07

Interactions between the prefrontal cortex (PFC) and basolateral amygdala (BLA) regulate emotional behaviors. However, a circuit-level understanding of functional connections between these brain regions remains incomplete. The BLA sends prominent glutamatergic projections to the PFC, but the overall influence of these inputs is predominantly inhibitory. Here we combine targeted recordings and optogenetics to examine the synaptic underpinnings of this inhibition in the mouse infralimbic PFC. We find that BLA inputs preferentially target layer 2 corticoamygdala over neighboring corticostriatal neurons. However, these inputs make even stronger connections onto neighboring parvalbumin and somatostatin expressing interneurons. Inhibitory connections from these two populations of interneurons are also much stronger onto corticoamygdala neurons. Consequently, BLA inputs are able to drive robust feedforward inhibition via two parallel interneuron pathways. Moreover, the contributions of these interneurons shift during repetitive activity, due to differences in short-term synaptic dynamics. Thus, parvalbumin interneurons are activated at the start of stimulus trains, whereas somatostatin interneuron activation builds during these trains. Together, these results reveal how the BLA impacts the PFC through a complex interplay of direct excitation and feedforward inhibition. They also highlight the roles of targeted connections onto multiple projection neurons and interneurons in this cortical circuit. Our findings provide a mechanistic understanding for how the BLA can influence the PFC circuit, with important implications for how this circuit participates in the regulation of emotion. The prefrontal cortex (PFC) and basolateral amygdala (BLA) interact to control emotional behaviors. Here we show that BLA inputs elicit direct excitation and feedforward inhibition of layer 2 projection neurons in infralimbic PFC. BLA inputs are much stronger at corticoamygdala neurons compared with nearby corticostriatal neurons. However, these inputs are even more powerful at parvalbumin and somatostatin expressing interneurons. BLA inputs thus activate two parallel inhibitory networks, whose contributions change during repetitive activity. Finally, connections from these interneurons are also more powerful at corticoamygdala neurons compared with corticostriatal neurons. Together, our results demonstrate how the BLA predominantly inhibits the PFC via a complex sequence involving multiple cell-type and input-specific connections. Copyright © 2016 the authors 0270-6474/16/369391-16$15.00/0.

Modelling and control of a nonlinear magnetostrictive actuator system

NASA Astrophysics Data System (ADS)

Ramli, M. H. M.; Majeed, A. P. P. Abdul; Anuar, M. A. M.; Mohamed, Z.

2018-04-01

This paper explores the implementation of a feedforward control method to a nonlinear control system, in particular, Magnetostrictive Actuators (MA) that has excellent properties of energy conversion between the mechanical and magnetic form through magnetostriction effects which could be used in actuating and sensing application. MA is known to exhibit hysteresis behaviour and it is rate dependent (the level of hysteresis depends closely on the rate of input excitation frequency). This is, nonetheless, an undesirable behaviour and has to be eliminated in realising high precision application. The MA is modelled by a phenomenological modelling approach via Prandtl-Ishlinskii (P-I) operator to characterise the hysteresis nonlinearities. A feedforward control strategy is designed and implemented to linearize and eliminate the hysteresis by model inversion. The results show that the P-I operator has the capability to model the hysteretic nonlinearity of MA with an acceptable accuracy. Furthermore, the proposed control scheme has demonstrated to be effective in providing superior trajectory tracking.

An artificial neural network prediction model of congenital heart disease based on risk factors: A hospital-based case-control study.

PubMed

Li, Huixia; Luo, Miyang; Zheng, Jianfei; Luo, Jiayou; Zeng, Rong; Feng, Na; Du, Qiyun; Fang, Junqun

2017-02-01

An artificial neural network (ANN) model was developed to predict the risks of congenital heart disease (CHD) in pregnant women.This hospital-based case-control study involved 119 CHD cases and 239 controls all recruited from birth defect surveillance hospitals in Hunan Province between July 2013 and June 2014. All subjects were interviewed face-to-face to fill in a questionnaire that covered 36 CHD-related variables. The 358 subjects were randomly divided into a training set and a testing set at the ratio of 85:15. The training set was used to identify the significant predictors of CHD by univariate logistic regression analyses and develop a standard feed-forward back-propagation neural network (BPNN) model for the prediction of CHD. The testing set was used to test and evaluate the performance of the ANN model. Univariate logistic regression analyses were performed on SPSS 18.0. The ANN models were developed on Matlab 7.1.The univariate logistic regression identified 15 predictors that were significantly associated with CHD, including education level (odds ratio  = 0.55), gravidity (1.95), parity (2.01), history of abnormal reproduction (2.49), family history of CHD (5.23), maternal chronic disease (4.19), maternal upper respiratory tract infection (2.08), environmental pollution around maternal dwelling place (3.63), maternal exposure to occupational hazards (3.53), maternal mental stress (2.48), paternal chronic disease (4.87), paternal exposure to occupational hazards (2.51), intake of vegetable/fruit (0.45), intake of fish/shrimp/meat/egg (0.59), and intake of milk/soymilk (0.55). After many trials, we selected a 3-layer BPNN model with 15, 12, and 1 neuron in the input, hidden, and output layers, respectively, as the best prediction model. The prediction model has accuracies of 0.91 and 0.86 on the training and testing sets, respectively. The sensitivity, specificity, and Yuden Index on the testing set (training set) are 0.78 (0.83), 0.90 (0.95), and 0.68 (0.78), respectively. The areas under the receiver operating curve on the testing and training sets are 0.87 and 0.97, respectively.This study suggests that the BPNN model could be used to predict the risk of CHD in individuals. This model should be further improved by large-sample-size research.

Predicting the spatial distribution of soil profile in Adapazari/Turkey by artificial neural networks using CPT data

NASA Astrophysics Data System (ADS)

Arel, Ersin

2012-06-01

The infamous soils of Adapazari, Turkey, that failed extensively during the 46-s long magnitude 7.4 earthquake in 1999 have since been the subject of a research program. Boreholes, piezocone soundings and voluminous laboratory testing have enabled researchers to apply sophisticated methods to determine the soil profiles in the city using the existing database. This paper describes the use of the artificial neural network (ANN) model to predict the complex soil profiles of Adapazari, based on cone penetration test (CPT) results. More than 3236 field CPT readings have been collected from 117 soundings spread over an area of 26 km2. An attempt has been made to develop the ANN model using multilayer perceptrons trained with a feed-forward back-propagation algorithm. The results show that the ANN model is fairly accurate in predicting complex soil profiles. Soil identification using CPT test results has principally been based on the Robertson charts. Applying neural network systems using the chart offers a powerful and rapid route to reliable prediction of the soil profiles.

Forecasting daily streamflow using online sequential extreme learning machines

NASA Astrophysics Data System (ADS)

Lima, Aranildo R.; Cannon, Alex J.; Hsieh, William W.

2016-06-01

While nonlinear machine methods have been widely used in environmental forecasting, in situations where new data arrive continually, the need to make frequent model updates can become cumbersome and computationally costly. To alleviate this problem, an online sequential learning algorithm for single hidden layer feedforward neural networks - the online sequential extreme learning machine (OSELM) - is automatically updated inexpensively as new data arrive (and the new data can then be discarded). OSELM was applied to forecast daily streamflow at two small watersheds in British Columbia, Canada, at lead times of 1-3 days. Predictors used were weather forecast data generated by the NOAA Global Ensemble Forecasting System (GEFS), and local hydro-meteorological observations. OSELM forecasts were tested with daily, monthly or yearly model updates. More frequent updating gave smaller forecast errors, including errors for data above the 90th percentile. Larger datasets used in the initial training of OSELM helped to find better parameters (number of hidden nodes) for the model, yielding better predictions. With the online sequential multiple linear regression (OSMLR) as benchmark, we concluded that OSELM is an attractive approach as it easily outperformed OSMLR in forecast accuracy.

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.

Small Body GN&C Research Report: A Robust Model Predictive Control Algorithm with Guaranteed Resolvability

NASA Technical Reports Server (NTRS)

Acikmese, Behcet A.; Carson, John M., III

2005-01-01

A robustly stabilizing MPC (model predictive control) algorithm for uncertain nonlinear systems is developed that guarantees the resolvability of the associated finite-horizon optimal control problem in a receding-horizon implementation. The control consists of two components; (i) feedforward, and (ii) feedback part. Feed-forward control is obtained by online solution of a finite-horizon optimal control problem for the nominal system dynamics. The feedback control policy is designed off-line based on a bound on the uncertainty in the system model. The entire controller is shown to be robustly stabilizing with a region of attraction composed of initial states for which the finite-horizon optimal control problem is feasible. The controller design for this algorithm is demonstrated on a class of systems with uncertain nonlinear terms that have norm-bounded derivatives, and derivatives in polytopes. An illustrative numerical example is also provided.

Seismic activity prediction using computational intelligence techniques in northern Pakistan

NASA Astrophysics Data System (ADS)

Asim, Khawaja M.; Awais, Muhammad; Martínez-Álvarez, F.; Iqbal, Talat

2017-10-01

Earthquake prediction study is carried out for the region of northern Pakistan. The prediction methodology includes interdisciplinary interaction of seismology and computational intelligence. Eight seismic parameters are computed based upon the past earthquakes. Predictive ability of these eight seismic parameters is evaluated in terms of information gain, which leads to the selection of six parameters to be used in prediction. Multiple computationally intelligent models have been developed for earthquake prediction using selected seismic parameters. These models include feed-forward neural network, recurrent neural network, random forest, multi layer perceptron, radial basis neural network, and support vector machine. The performance of every prediction model is evaluated and McNemar's statistical test is applied to observe the statistical significance of computational methodologies. Feed-forward neural network shows statistically significant predictions along with accuracy of 75% and positive predictive value of 78% in context of northern Pakistan.

A robust model predictive control algorithm for uncertain nonlinear systems that guarantees resolvability

NASA Technical Reports Server (NTRS)

Acikmese, Ahmet Behcet; Carson, John M., III

2006-01-01

A robustly stabilizing MPC (model predictive control) algorithm for uncertain nonlinear systems is developed that guarantees resolvability. With resolvability, initial feasibility of the finite-horizon optimal control problem implies future feasibility in a receding-horizon framework. The control consists of two components; (i) feed-forward, and (ii) feedback part. Feed-forward control is obtained by online solution of a finite-horizon optimal control problem for the nominal system dynamics. The feedback control policy is designed off-line based on a bound on the uncertainty in the system model. The entire controller is shown to be robustly stabilizing with a region of attraction composed of initial states for which the finite-horizon optimal control problem is feasible. The controller design for this algorithm is demonstrated on a class of systems with uncertain nonlinear terms that have norm-bounded derivatives and derivatives in polytopes. An illustrative numerical example is also provided.

Direct adaptive control of manipulators in Cartesian space

NASA Technical Reports Server (NTRS)

Seraji, H.

1987-01-01

A new adaptive-control scheme for direct control of manipulator end effector to achieve trajectory tracking in Cartesian space is developed in this article. The control structure is obtained from linear multivariable theory and is composed of simple feedforward and feedback controllers and an auxiliary input. The direct adaptation laws are derived from model reference adaptive control theory and are not based on parameter estimation of the robot model. The utilization of adaptive feedforward control and the inclusion of auxiliary input are novel features of the present scheme and result in improved dynamic performance over existing adaptive control schemes. The adaptive controller does not require the complex mathematical model of the robot dynamics or any knowledge of the robot parameters or the payload, and is computationally fast for on-line implementation with high sampling rates. The control scheme is applied to a two-link manipulator for illustration.

Comparing taxi clearance input layouts for advancements in flight deck automation for surface operations

NASA Astrophysics Data System (ADS)

Cheng, Lara W. S.

Airport moving maps (AMMs) have been shown to decrease navigation errors, increase taxiing speed, and reduce workload when they depict airport layout, current aircraft position, and the cleared taxi route. However, current technologies are limited in their ability to depict the cleared taxi route due to the unavailability of datacomm or other means of electronically transmitting clearances from ATC to the flight deck. This study examined methods by which pilots can input ATC-issued taxi clearances to support taxi route depictions on the AMM. Sixteen general aviation (GA) pilots used a touchscreen monitor to input taxi clearances using two input layouts, softkeys and QWERTY, each with and without feedforward (graying out invalid inputs). QWERTY yielded more taxi route input errors than the softkeys layout. The presence of feedforward did not produce fewer taxi route input errors than in the non-feedforward condition. The QWERTY layout did reduce taxi clearance input times relative to the softkeys layout, but when feedforward was present this effect was observed only for the longer, 6-segment taxi clearances. It was observed that with the softkeys layout, feedforward reduced input times compared to non-feedforward but only for the 4-segment clearances. Feedforward did not support faster taxi clearance input times for the QWERTY layout. Based on the results and analyses of the present study, it is concluded that for taxi clearance inputs, (1) QWERTY remain the standard for alphanumeric inputs, and (2) feedforward be investigated further, with a focus on participant preference and performance of black-gray contrast of keys.

A combined stochastic feedforward and feedback control design methodology with application to autoland design

NASA Technical Reports Server (NTRS)

Halyo, Nesim

1987-01-01

A combined stochastic feedforward and feedback control design methodology was developed. The objective of the feedforward control law is to track the commanded trajectory, whereas the feedback control law tries to maintain the plant state near the desired trajectory in the presence of disturbances and uncertainties about the plant. The feedforward control law design is formulated as a stochastic optimization problem and is embedded into the stochastic output feedback problem where the plant contains unstable and uncontrollable modes. An algorithm to compute the optimal feedforward is developed. In this approach, the use of error integral feedback, dynamic compensation, control rate command structures are an integral part of the methodology. An incremental implementation is recommended. Results on the eigenvalues of the implemented versus designed control laws are presented. The stochastic feedforward/feedback control methodology is used to design a digital automatic landing system for the ATOPS Research Vehicle, a Boeing 737-100 aircraft. The system control modes include localizer and glideslope capture and track, and flare to touchdown. Results of a detailed nonlinear simulation of the digital control laws, actuator systems, and aircraft aerodynamics are presented.

Artificial neural network for suppression of banding artifacts in balanced steady-state free precession MRI.

PubMed

Kim, Ki Hwan; Park, Sung-Hong

2017-04-01

The balanced steady-state free precession (bSSFP) MR sequence is frequently used in clinics, but is sensitive to off-resonance effects, which can cause banding artifacts. Often multiple bSSFP datasets are acquired at different phase cycling (PC) angles and then combined in a special way for banding artifact suppression. Many strategies of combining the datasets have been suggested for banding artifact suppression, but there are still limitations in their performance, especially when the number of phase-cycled bSSFP datasets is small. The purpose of this study is to develop a learning-based model to combine the multiple phase-cycled bSSFP datasets for better banding artifact suppression. Multilayer perceptron (MLP) is a feedforward artificial neural network consisting of three layers of input, hidden, and output layers. MLP models were trained by input bSSFP datasets acquired from human brain and knee at 3T, which were separately performed for two and four PC angles. Banding-free bSSFP images were generated by maximum-intensity projection (MIP) of 8 or 12 phase-cycled datasets and were used as targets for training the output layer. The trained MLP models were applied to another brain and knee datasets acquired with different scan parameters and also to multiple phase-cycled bSSFP functional MRI datasets acquired on rat brain at 9.4T, in comparison with the conventional MIP method. Simulations were also performed to validate the MLP approach. Both the simulations and human experiments demonstrated that MLP suppressed banding artifacts significantly, superior to MIP in both banding artifact suppression and SNR efficiency. MLP demonstrated superior performance over MIP for the 9.4T fMRI data as well, which was not used for training the models, while visually preserving the fMRI maps very well. Artificial neural network is a promising technique for combining multiple phase-cycled bSSFP datasets for banding artifact suppression. Copyright © 2016 Elsevier Inc. All rights reserved.

Top-level dynamics and the regulated gene response of feed-forward loop transcriptional motifs.

PubMed

Mayo, Michael; Abdelzaher, Ahmed; Perkins, Edward J; Ghosh, Preetam

2014-09-01

Feed-forward loops are hierarchical three-node transcriptional subnetworks, wherein a top-level protein regulates the activity of a target gene via two paths: a direct-regulatory path, and an indirect route, whereby the top-level proteins act implicitly through an intermediate transcription factor. Using a transcriptional network of the model bacterium Escherichia coli, we confirmed that nearly all types of feed-forward loop were significantly overrepresented in the bacterial network. We then used mathematical modeling to study their dynamics by manipulating the rise times of the top-level protein concentration, termed the induction time, through alteration of the protein destruction rates. Rise times of the regulated proteins exhibited two qualitatively different regimes, depending on whether top-level inductions were "fast" or "slow." In the fast regime, rise times were nearly independent of rapid top-level inductions, indicative of biological robustness, and occurred when RNA production rate-limits the protein yield. Alternatively, the protein rise times were dependent upon slower top-level inductions, greater than approximately one bacterial cell cycle. An equation is given for this crossover, which depends upon three parameters of the direct-regulatory path: transcriptional cooperation at the DNA-binding site, a protein-DNA dissociation constant, and the relative magnitude of the top-level protien concentration.

Quantitative workflow based on NN for weighting criteria in landfill suitability mapping

NASA Astrophysics Data System (ADS)

Abujayyab, Sohaib K. M.; Ahamad, Mohd Sanusi S.; Yahya, Ahmad Shukri; Ahmad, Siti Zubaidah; Alkhasawneh, Mutasem Sh.; Aziz, Hamidi Abdul

2017-10-01

Our study aims to introduce a new quantitative workflow that integrates neural networks (NNs) and multi criteria decision analysis (MCDA). Existing MCDA workflows reveal a number of drawbacks, because of the reliance on human knowledge in the weighting stage. Thus, new workflow presented to form suitability maps at the regional scale for solid waste planning based on NNs. A feed-forward neural network employed in the workflow. A total of 34 criteria were pre-processed to establish the input dataset for NN modelling. The final learned network used to acquire the weights of the criteria. Accuracies of 95.2% and 93.2% achieved for the training dataset and testing dataset, respectively. The workflow was found to be capable of reducing human interference to generate highly reliable maps. The proposed workflow reveals the applicability of NN in generating landfill suitability maps and the feasibility of integrating them with existing MCDA workflows.

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.

TRANSIENT BEAM LOADING EFFECTS IN RF SYSTEMS IN JLEIC

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

Wang, Haipeng; Guo, Jiquan; Rimmer, Robert A.

2016-05-01

The pulsed electron bunch trains generated from the Continuous Electron Beam Accelerator Facility (CEBAF) linac to inject into the proposed Jefferson Lab Electron Ion Collider (JLEIC) e-ring will produce transient beam loading effects in the Superconducting Radio Frequency (SRF) systems that, if not mitigated, could cause unacceptably large beam energy deviation in the injection capture, or exceed the energy acceptance of CEBAF’s recirculating arcs. In the electron storage ring, the beam abort or ion clearing gaps or uneven bucket filling can cause large beam phase transients in the (S)RF cavity control systems and even beam loss due to Robinson instability.more » We have first analysed the beam stability criteria in steady state and estimated the transient effect in Feedforward and Feedback RF controls. Initial analytical models for these effects are shown for the design of the JLEIC e-ring from 3GeV to 12GeV.« less

Approximating quantum many-body wave functions using artificial neural networks

NASA Astrophysics Data System (ADS)

Cai, Zi; Liu, Jinguo

2018-01-01

In this paper, we demonstrate the expressibility of artificial neural networks (ANNs) in quantum many-body physics by showing that a feed-forward neural network with a small number of hidden layers can be trained to approximate with high precision the ground states of some notable quantum many-body systems. We consider the one-dimensional free bosons and fermions, spinless fermions on a square lattice away from half-filling, as well as frustrated quantum magnetism with a rapidly oscillating ground-state characteristic function. In the latter case, an ANN with a standard architecture fails, while that with a slightly modified one successfully learns the frustration-induced complex sign rule in the ground state and approximates the ground states with high precisions. As an example of practical use of our method, we also perform the variational method to explore the ground state of an antiferromagnetic J1-J2 Heisenberg model.

Learning the Relationship between the Primary Structure of HIV Envelope Glycoproteins and Neutralization Activity of Particular Antibodies by Using Artificial Neural Networks

PubMed Central

Buiu, Cătălin; Putz, Mihai V.; Avram, Speranta

2016-01-01

The dependency between the primary structure of HIV envelope glycoproteins (ENV) and the neutralization data for given antibodies is very complicated and depends on a large number of factors, such as the binding affinity of a given antibody for a given ENV protein, and the intrinsic infection kinetics of the viral strain. This paper presents a first approach to learning these dependencies using an artificial feedforward neural network which is trained to learn from experimental data. The results presented here demonstrate that the trained neural network is able to generalize on new viral strains and to predict reliable values of neutralizing activities of given antibodies against HIV-1. PMID:27727189

Artificial neural networks in Space Station optimal attitude control

NASA Astrophysics Data System (ADS)

Kumar, Renjith R.; Seywald, Hans; Deshpande, Samir M.; Rahman, Zia

1992-08-01

Innovative techniques of using 'Artificial Neural Networks' (ANN) for improving the performance of the pitch axis attitude control system of Space Station Freedom using Control Moment Gyros (CMGs) are investigated. The first technique uses a feedforward ANN with multilayer perceptrons to obtain an on-line controller which improves the performance of the control system via a model following approach. The second techique uses a single layer feedforward ANN with a modified back propagation scheme to estimate the internal plant variations and the external disturbances separately. These estimates are then used to solve two differential Riccati equations to obtain time varying gains which improve the control system performance in successive orbits.

Artificial neural networks for defining the water quality determinants of groundwater abstraction in coastal aquifer

NASA Astrophysics Data System (ADS)

Lallahem, S.; Hani, A.

2017-02-01

Water sustainability in the lower Seybouse River basin, eastern Algeria, must take into account the importance of water quantity and quality integration. So, there is a need for a better knowledge and understanding of the water quality determinants of groundwater abstraction to meet the municipal and agricultural uses. In this paper, the artificial neural network (ANN) models were used to model and predict the relationship between groundwater abstraction and water quality determinants in the lower Seybouse River basin. The study area chosen is the lower Seybouse River basin and real data were collected from forty five wells for reference year 2006. Results indicate that the feed-forward multilayer perceptron models with back-propagation are useful tools to define and prioritize the important water quality parameters of groundwater abstraction and use. The model evaluation shows that the correlation coefficients are more than 95% for training, verification and testing data. The model aims to link the water quantity and quality with the objective to strengthen the Integrated Water Resources Management approach. It assists water planners and managers to better assess the water quality parameters and progress towards the provision of appropriate quantities of water of suitable quality.

Neural network application to aircraft control system design

NASA Technical Reports Server (NTRS)

Troudet, Terry; Garg, Sanjay; Merrill, Walter C.

1991-01-01

The feasibility of using artificial neural networks as control systems for modern, complex aerospace vehicles is investigated via an example aircraft control design study. The problem considered is that of designing a controller for an integrated airframe/propulsion longitudinal dynamics model of a modern fighter aircraft to provide independent control of pitch rate and airspeed responses to pilot command inputs. An explicit model following controller using H infinity control design techniques is first designed to gain insight into the control problem as well as to provide a baseline for evaluation of the neurocontroller. Using the model of the desired dynamics as a command generator, a multilayer feedforward neural network is trained to control the vehicle model within the physical limitations of the actuator dynamics. This is achieved by minimizing an objective function which is a weighted sum of tracking errors and control input commands and rates. To gain insight in the neurocontrol, linearized representations of the nonlinear neurocontroller are analyzed along a commanded trajectory. Linear robustness analysis tools are then applied to the linearized neurocontroller models and to the baseline H infinity based controller. Future areas of research are identified to enhance the practical applicability of neural networks to flight control design.

Neural network application to aircraft control system design

NASA Technical Reports Server (NTRS)

Troudet, Terry; Garg, Sanjay; Merrill, Walter C.

1991-01-01

The feasibility of using artificial neural network as control systems for modern, complex aerospace vehicles is investigated via an example aircraft control design study. The problem considered is that of designing a controller for an integrated airframe/propulsion longitudinal dynamics model of a modern fighter aircraft to provide independent control of pitch rate and airspeed responses to pilot command inputs. An explicit model following controller using H infinity control design techniques is first designed to gain insight into the control problem as well as to provide a baseline for evaluation of the neurocontroller. Using the model of the desired dynamics as a command generator, a multilayer feedforward neural network is trained to control the vehicle model within the physical limitations of the actuator dynamics. This is achieved by minimizing an objective function which is a weighted sum of tracking errors and control input commands and rates. To gain insight in the neurocontrol, linearized representations of the nonlinear neurocontroller are analyzed along a commanded trajectory. Linear robustness analysis tools are then applied to the linearized neurocontroller models and to the baseline H infinity based controller. Future areas of research identified to enhance the practical applicability of neural networks to flight control design.

Application of principal component regression and artificial neural network in FT-NIR soluble solids content determination of intact pear fruit

NASA Astrophysics Data System (ADS)

Ying, Yibin; Liu, Yande; Fu, Xiaping; Lu, Huishan

2005-11-01

The artificial neural networks (ANNs) have been used successfully in applications such as pattern recognition, image processing, automation and control. However, majority of today's applications of ANNs is back-propagate feed-forward ANN (BP-ANN). In this paper, back-propagation artificial neural networks (BP-ANN) were applied for modeling soluble solid content (SSC) of intact pear from their Fourier transform near infrared (FT-NIR) spectra. One hundred and sixty-four pear samples were used to build the calibration models and evaluate the models predictive ability. The results are compared to the classical calibration approaches, i.e. principal component regression (PCR), partial least squares (PLS) and non-linear PLS (NPLS). The effects of the optimal methods of training parameters on the prediction model were also investigated. BP-ANN combine with principle component regression (PCR) resulted always better than the classical PCR, PLS and Weight-PLS methods, from the point of view of the predictive ability. Based on the results, it can be concluded that FT-NIR spectroscopy and BP-ANN models can be properly employed for rapid and nondestructive determination of fruit internal quality.

Impaired Feedforward Control and Enhanced Feedback Control of Speech in Patients with Cerebellar Degeneration

PubMed Central

Agnew, Zarinah; Nagarajan, Srikantan; Houde, John; Ivry, Richard B.

2017-01-01

The cerebellum has been hypothesized to form a crucial part of the speech motor control network. Evidence for this comes from patients with cerebellar damage, who exhibit a variety of speech deficits, as well as imaging studies showing cerebellar activation during speech production in healthy individuals. To date, the precise role of the cerebellum in speech motor control remains unclear, as it has been implicated in both anticipatory (feedforward) and reactive (feedback) control. Here, we assess both anticipatory and reactive aspects of speech motor control, comparing the performance of patients with cerebellar degeneration and matched controls. Experiment 1 tested feedforward control by examining speech adaptation across trials in response to a consistent perturbation of auditory feedback. Experiment 2 tested feedback control, examining online corrections in response to inconsistent perturbations of auditory feedback. Both male and female patients and controls were tested. The patients were impaired in adapting their feedforward control system relative to controls, exhibiting an attenuated anticipatory response to the perturbation. In contrast, the patients produced even larger compensatory responses than controls, suggesting an increased reliance on sensory feedback to guide speech articulation in this population. Together, these results suggest that the cerebellum is crucial for maintaining accurate feedforward control of speech, but relatively uninvolved in feedback control. SIGNIFICANCE STATEMENT Speech motor control is a complex activity that is thought to rely on both predictive, feedforward control as well as reactive, feedback control. While the cerebellum has been shown to be part of the speech motor control network, its functional contribution to feedback and feedforward control remains controversial. Here, we use real-time auditory perturbations of speech to show that patients with cerebellar degeneration are impaired in adapting feedforward control of speech but retain the ability to make online feedback corrections; indeed, the patients show an increased sensitivity to feedback. These results indicate that the cerebellum forms a crucial part of the feedforward control system for speech but is not essential for online, feedback control. PMID:28842410

Using oceanic-atmospheric oscillations for long lead time streamflow forecasting

NASA Astrophysics Data System (ADS)

Kalra, Ajay; Ahmad, Sajjad

2009-03-01

We present a data-driven model, Support Vector Machine (SVM), for long lead time streamflow forecasting using oceanic-atmospheric oscillations. The SVM is based on statistical learning theory that uses a hypothesis space of linear functions based on Kernel approach and has been used to predict a quantity forward in time on the basis of training from past data. The strength of SVM lies in minimizing the empirical classification error and maximizing the geometric margin by solving inverse problem. The SVM model is applied to three gages, i.e., Cisco, Green River, and Lees Ferry in the Upper Colorado River Basin in the western United States. Annual oceanic-atmospheric indices, comprising Pacific Decadal Oscillation (PDO), North Atlantic Oscillation (NAO), Atlantic Multidecadal Oscillation (AMO), and El Nino-Southern Oscillations (ENSO) for a period of 1906-2001 are used to generate annual streamflow volumes with 3 years lead time. The SVM model is trained with 86 years of data (1906-1991) and tested with 10 years of data (1992-2001). On the basis of correlation coefficient, root means square error, and Nash Sutcliffe Efficiency Coefficient the model shows satisfactory results, and the predictions are in good agreement with measured streamflow volumes. Sensitivity analysis, performed to evaluate the effect of individual and coupled oscillations, reveals a strong signal for ENSO and NAO indices as compared to PDO and AMO indices for the long lead time streamflow forecast. Streamflow predictions from the SVM model are found to be better when compared with the predictions obtained from feedforward back propagation artificial neural network model and linear regression.

A spherical model for orientation and spatial-frequency tuning in a cortical hypercolumn.

PubMed Central

Bressloff, Paul C; Cowan, Jack D

2003-01-01

A theory is presented of the way in which the hypercolumns in primary visual cortex (V1) are organized to detect important features of visual images, namely local orientation and spatial-frequency. Given the existence in V1 of dual maps for these features, both organized around orientation pinwheels, we constructed a model of a hypercolumn in which orientation and spatial-frequency preferences are represented by the two angular coordinates of a sphere. The two poles of this sphere are taken to correspond, respectively, to high and low spatial-frequency preferences. In Part I of the paper, we use mean-field methods to derive exact solutions for localized activity states on the sphere. We show how cortical amplification through recurrent interactions generates a sharply tuned, contrast-invariant population response to both local orientation and local spatial frequency, even in the case of a weakly biased input from the lateral geniculate nucleus (LGN). A major prediction of our model is that this response is non-separable with respect to the local orientation and spatial frequency of a stimulus. That is, orientation tuning is weaker around the pinwheels, and there is a shift in spatial-frequency tuning towards that of the closest pinwheel at non-optimal orientations. In Part II of the paper, we demonstrate that a simple feed-forward model of spatial-frequency preference, unlike that for orientation preference, does not generate a faithful representation when amplified by recurrent interactions in V1. We then introduce the idea that cortico-geniculate feedback modulates LGN activity to generate a faithful representation, thus providing a new functional interpretation of the role of this feedback pathway. Using linear filter theory, we show that if the feedback from a cortical cell is taken to be approximately equal to the reciprocal of the corresponding feed-forward receptive field (in the two-dimensional Fourier domain), then the mismatch between the feed-forward and cortical frequency representations is eliminated. We therefore predict that cortico-geniculate feedback connections innervate the LGN in a pattern determined by the orientation and spatial-frequency biases of feed-forward receptive fields. Finally, we show how recurrent cortical interactions can generate cross-orientation suppression. PMID:14561324

Erosion and refilling of the plasmasphere during a geomagnetic storm modeled by a neural network

NASA Astrophysics Data System (ADS)

Chu, X. N.; Bortnik, J.; Li, W.; Ma, Q.; Angelopoulos, V.; Thorne, R. M.

2017-07-01

We present a history-dependent model of the equatorial plasma density of the inner magnetosphere using a feedforward neural network with two hidden layers. As the model inputs, we take locations and time series of SYM-H, AL, and F10.7 indices. By considering not only the instantaneous values but also the past values of geomagnetic and solar indices, the model is history dependent on levels of geomagnetic and solar activity. The modeled electron density is continuous both spatially and temporally so that the evolution of the density can be studied (such as plasmaspheric refilling). The model is trained using the electron density inferred from the spacecraft potential from three THEMIS probes. The equatorial electron density is shown to be accurately reconstructed with a correlation coefficient of r 0.953 between data and model target. Since the model is history dependent, it succeeds in reconstructing various density features and dynamic behaviors, such as the quiet time plasmasphere, erosion and recovery of the plasmasphere, as well as the plume formation during a storm on 4 February 2011. Our model may provide unprecedented insight into the behavior of the equatorial density at any time and location; as an example we show the inferred refilling rate from our model and compare it to previous estimates.

Further results on open-loop compensation of rate-dependent hysteresis in a magnetostrictive actuator with the Prandtl-Ishlinskii model

NASA Astrophysics Data System (ADS)

Al Janaideh, Mohammad; Aljanaideh, Omar

2018-05-01

Apart from the output-input hysteresis loops, the magnetostrictive actuators also exhibit asymmetry and saturation, particularly under moderate to large magnitude inputs and at relatively higher frequencies. Such nonlinear input-output characteristics could be effectively characterized by a rate-dependent Prandtl-Ishlinskii model in conjunction with a function of deadband operators. In this study, an inverse model is formulated to seek real-time compensation of rate-dependent and asymmetric hysteresis nonlinearities of a Terfenol-D magnetostrictive actuator. The inverse model is formulated with the inverse of the rate-dependent Prandtl-Ishlinskii model, satisfying the threshold dilation condition, with the inverse of the deadband function. The inverse model was subsequently applied to the hysteresis model as a feedforward compensator. The proposed compensator is applied as a feedforward compensator to the actuator hardware to study its potential for rate-dependent and asymmetric hysteresis loops. The experimental results are obtained under harmonic and complex harmonic inputs further revealed that the inverse compensator can substantially suppress the hysteresis and output asymmetry nonlinearities in the entire frequency range considered in the study.

Generation of the Human Biped Stance by a Neural Controller Able to Compensate Neurological Time Delay

PubMed Central

Jiang, Ping; Chiba, Ryosuke; Takakusaki, Kaoru; Ota, Jun

2016-01-01

The development of a physiologically plausible computational model of a neural controller that can realize a human-like biped stance is important for a large number of potential applications, such as assisting device development and designing robotic control systems. In this paper, we develop a computational model of a neural controller that can maintain a musculoskeletal model in a standing position, while incorporating a 120-ms neurological time delay. Unlike previous studies that have used an inverted pendulum model, a musculoskeletal model with seven joints and 70 muscular-tendon actuators is adopted to represent the human anatomy. Our proposed neural controller is composed of both feed-forward and feedback controls. The feed-forward control corresponds to the constant activation input necessary for the musculoskeletal model to maintain a standing posture. This compensates for gravity and regulates stiffness. The developed neural controller model can replicate two salient features of the human biped stance: (1) physiologically plausible muscle activations for quiet standing; and (2) selection of a low active stiffness for low energy consumption. PMID:27655271

Deep neural networks show an equivalent and often superior performance to dermatologists in onychomycosis diagnosis: Automatic construction of onychomycosis datasets by region-based convolutional deep neural network.

PubMed

Han, Seung Seog; Park, Gyeong Hun; Lim, Woohyung; Kim, Myoung Shin; Na, Jung Im; Park, Ilwoo; Chang, Sung Eun

2018-01-01

Although there have been reports of the successful diagnosis of skin disorders using deep learning, unrealistically large clinical image datasets are required for artificial intelligence (AI) training. We created datasets of standardized nail images using a region-based convolutional neural network (R-CNN) trained to distinguish the nail from the background. We used R-CNN to generate training datasets of 49,567 images, which we then used to fine-tune the ResNet-152 and VGG-19 models. The validation datasets comprised 100 and 194 images from Inje University (B1 and B2 datasets, respectively), 125 images from Hallym University (C dataset), and 939 images from Seoul National University (D dataset). The AI (ensemble model; ResNet-152 + VGG-19 + feedforward neural networks) results showed test sensitivity/specificity/ area under the curve values of (96.0 / 94.7 / 0.98), (82.7 / 96.7 / 0.95), (92.3 / 79.3 / 0.93), (87.7 / 69.3 / 0.82) for the B1, B2, C, and D datasets. With a combination of the B1 and C datasets, the AI Youden index was significantly (p = 0.01) higher than that of 42 dermatologists doing the same assessment manually. For B1+C and B2+ D dataset combinations, almost none of the dermatologists performed as well as the AI. By training with a dataset comprising 49,567 images, we achieved a diagnostic accuracy for onychomycosis using deep learning that was superior to that of most of the dermatologists who participated in this study.

A hybrid deep neural network and physically based distributed model for river stage prediction

NASA Astrophysics Data System (ADS)

hitokoto, Masayuki; sakuraba, Masaaki

2016-04-01

We developed the real-time river stage prediction model, using the hybrid deep neural network and physically based distributed model. As the basic model, 4 layer feed-forward artificial neural network (ANN) was used. As a network training method, the deep learning technique was applied. To optimize the network weight, the stochastic gradient descent method based on the back propagation method was used. As a pre-training method, the denoising autoencoder was used. Input of the ANN model is hourly change of water level and hourly rainfall, output data is water level of downstream station. In general, the desirable input of the ANN has strong correlation with the output. In conceptual hydrological model such as tank model and storage-function model, river discharge is governed by the catchment storage. Therefore, the change of the catchment storage, downstream discharge subtracted from rainfall, can be the potent input candidate of the ANN model instead of rainfall. From this point of view, the hybrid deep neural network and physically based distributed model was developed. The prediction procedure of the hybrid model is as follows; first, downstream discharge was calculated by the distributed model, and then estimates the hourly change of catchment storage form rainfall and calculated discharge as the input of the ANN model, and finally the ANN model was calculated. In the training phase, hourly change of catchment storage can be calculated by the observed rainfall and discharge data. The developed model was applied to the one catchment of the OOYODO River, one of the first-grade river in Japan. The modeled catchment is 695 square km. For the training data, 5 water level gauging station and 14 rain-gauge station in the catchment was used. The training floods, superior 24 events, were selected during the period of 2005-2014. Prediction was made up to 6 hours, and 6 models were developed for each prediction time. To set the proper learning parameters and network architecture of the ANN model, sensitivity analysis was done by the case study approach. The prediction result was evaluated by the superior 4 flood events by the leave-one-out cross validation. The prediction result of the basic 4 layer ANN was better than the conventional 3 layer ANN model. However, the result did not reproduce well the biggest flood event, supposedly because the lack of the sufficient high-water level flood event in the training data. The result of the hybrid model outperforms the basic ANN model and distributed model, especially improved the performance of the basic ANN model in the biggest flood event.

Feed-Forward: Students Gaining More from Assessment via Deeper Engagement in Video-Recorded Presentations

ERIC Educational Resources Information Center

Murphy, Karen; Barry, Shane

2016-01-01

Presentation feedback can be limited in its feed-forward value, as students do not have their actual presentation available for review whilst reflecting upon the feedback. This study reports on students' perceptions of the learning and feed-forward value of an oral presentation assessment. Students self-marked their performance immediately after…

Perceptual learning increases the strength of the earliest signals in visual cortex.

PubMed

Bao, Min; Yang, Lin; Rios, Cristina; He, Bin; Engel, Stephen A

2010-11-10

Training improves performance on most visual tasks. Such perceptual learning can modify how information is read out from, and represented in, later visual areas, but effects on early visual cortex are controversial. In particular, it remains unknown whether learning can reshape neural response properties in early visual areas independent from feedback arising in later cortical areas. Here, we tested whether learning can modify feedforward signals in early visual cortex as measured by the human electroencephalogram. Fourteen subjects were trained for >24 d to detect a diagonal grating pattern in one quadrant of the visual field. Training improved performance, reducing the contrast needed for reliable detection, and also reliably increased the amplitude of the earliest component of the visual evoked potential, the C1. Control orientations and locations showed smaller effects of training. Because the C1 arises rapidly and has a source in early visual cortex, our results suggest that learning can increase early visual area response through local receptive field changes without feedback from later areas.

Feature Extraction of Event-Related Potentials Using Wavelets: An Application to Human Performance Monitoring

NASA Technical Reports Server (NTRS)

Trejo, Leonard J.; Shensa, Mark J.; Remington, Roger W. (Technical Monitor)

1998-01-01

This report describes the development and evaluation of mathematical models for predicting human performance from discrete wavelet transforms (DWT) of event-related potentials (ERP) elicited by task-relevant stimuli. The DWT was compared to principal components analysis (PCA) for representation of ERPs in linear regression and neural network models developed to predict a composite measure of human signal detection performance. Linear regression models based on coefficients of the decimated DWT predicted signal detection performance with half as many f ree parameters as comparable models based on PCA scores. In addition, the DWT-based models were more resistant to model degradation due to over-fitting than PCA-based models. Feed-forward neural networks were trained using the backpropagation,-, algorithm to predict signal detection performance based on raw ERPs, PCA scores, or high-power coefficients of the DWT. Neural networks based on high-power DWT coefficients trained with fewer iterations, generalized to new data better, and were more resistant to overfitting than networks based on raw ERPs. Networks based on PCA scores did not generalize to new data as well as either the DWT network or the raw ERP network. The results show that wavelet expansions represent the ERP efficiently and extract behaviorally important features for use in linear regression or neural network models of human performance. The efficiency of the DWT is discussed in terms of its decorrelation and energy compaction properties. In addition, the DWT models provided evidence that a pattern of low-frequency activity (1 to 3.5 Hz) occurring at specific times and scalp locations is a reliable correlate of human signal detection performance.

Feature extraction of event-related potentials using wavelets: an application to human performance monitoring

NASA Technical Reports Server (NTRS)

Trejo, L. J.; Shensa, M. J.

1999-01-01

This report describes the development and evaluation of mathematical models for predicting human performance from discrete wavelet transforms (DWT) of event-related potentials (ERP) elicited by task-relevant stimuli. The DWT was compared to principal components analysis (PCA) for representation of ERPs in linear regression and neural network models developed to predict a composite measure of human signal detection performance. Linear regression models based on coefficients of the decimated DWT predicted signal detection performance with half as many free parameters as comparable models based on PCA scores. In addition, the DWT-based models were more resistant to model degradation due to over-fitting than PCA-based models. Feed-forward neural networks were trained using the backpropagation algorithm to predict signal detection performance based on raw ERPs, PCA scores, or high-power coefficients of the DWT. Neural networks based on high-power DWT coefficients trained with fewer iterations, generalized to new data better, and were more resistant to overfitting than networks based on raw ERPs. Networks based on PCA scores did not generalize to new data as well as either the DWT network or the raw ERP network. The results show that wavelet expansions represent the ERP efficiently and extract behaviorally important features for use in linear regression or neural network models of human performance. The efficiency of the DWT is discussed in terms of its decorrelation and energy compaction properties. In addition, the DWT models provided evidence that a pattern of low-frequency activity (1 to 3.5 Hz) occurring at specific times and scalp locations is a reliable correlate of human signal detection performance. Copyright 1999 Academic Press.

Artificial intelligence modeling of cadmium(II) biosorption using rice straw

NASA Astrophysics Data System (ADS)

Nasr, Mahmoud; Mahmoud, Alaa El Din; Fawzy, Manal; Radwan, Ahmed

2017-05-01

The biosorption efficiency of Cd2+ using rice straw was investigated at room temperature (25 ± 4 °C), contact time (2 h) and agitation rate (5 Hz). Experiments studied the effect of three factors, biosorbent dose BD (0.1 and 0.5 g/L), pH (2 and 7) and initial Cd2+ concentration X (10 and 100 mg/L) at two levels "low" and "high". Results showed that, a variation in X from high to low revealed 31 % increase in the Cd2+ biosorption. However, a discrepancy in pH and BD from low to high achieved 28.60 and 23.61 % increase in the removal of Cd2+, respectively. From 23 factorial design, the effects of BD, pH and X achieved p value equals to 0.2248, 0.1881 and 0.1742, respectively, indicating that the influences are in the order X > pH > BD. Similarly, an adaptive neuro-fuzzy inference system indicated that X is the most influential with training and checking errors of 10.87 and 17.94, respectively. This trend was followed by "pH" with training error (15.80) and checking error (17.39), after that BD with training error (16.09) and checking error (16.29). A feed-forward back-propagation neural network with a configuration 3-6-1 achieved correlation ( R) of 0.99 (training), 0.82 (validation) and 0.97 (testing). Thus, the proposed network is capable of predicting Cd2+ biosorption with high accuracy, while the most significant variable was X.

Development and Validation of a Deep Neural Network Model for Prediction of Postoperative In-hospital Mortality.

PubMed

Lee, Christine K; Hofer, Ira; Gabel, Eilon; Baldi, Pierre; Cannesson, Maxime

2018-04-17

The authors tested the hypothesis that deep neural networks trained on intraoperative features can predict postoperative in-hospital mortality. The data used to train and validate the algorithm consists of 59,985 patients with 87 features extracted at the end of surgery. Feed-forward networks with a logistic output were trained using stochastic gradient descent with momentum. The deep neural networks were trained on 80% of the data, with 20% reserved for testing. The authors assessed improvement of the deep neural network by adding American Society of Anesthesiologists (ASA) Physical Status Classification and robustness of the deep neural network to a reduced feature set. The networks were then compared to ASA Physical Status, logistic regression, and other published clinical scores including the Surgical Apgar, Preoperative Score to Predict Postoperative Mortality, Risk Quantification Index, and the Risk Stratification Index. In-hospital mortality in the training and test sets were 0.81% and 0.73%. The deep neural network with a reduced feature set and ASA Physical Status classification had the highest area under the receiver operating characteristics curve, 0.91 (95% CI, 0.88 to 0.93). The highest logistic regression area under the curve was found with a reduced feature set and ASA Physical Status (0.90, 95% CI, 0.87 to 0.93). The Risk Stratification Index had the highest area under the receiver operating characteristics curve, at 0.97 (95% CI, 0.94 to 0.99). Deep neural networks can predict in-hospital mortality based on automatically extractable intraoperative data, but are not (yet) superior to existing methods.

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.

Infrared differential-absorption Mueller matrix spectroscopy and neural network-based data fusion for biological aerosol standoff detection.

PubMed

Carrieri, Arthur H; Copper, Jack; Owens, David J; Roese, Erik S; Bottiger, Jerold R; Everly, Robert D; Hung, Kevin C

2010-01-20

An active spectrophotopolarimeter sensor and support system were developed for a military/civilian defense feasibility study concerning the identification and standoff detection of biological aerosols. Plumes of warfare agent surrogates gamma-irradiated Bacillus subtilis and chicken egg white albumen (analytes), Arizona road dust (terrestrial interferent), water mist (atmospheric interferent), and talcum powders (experiment controls) were dispersed inside windowless chambers and interrogated by multiple CO(2) laser beams spanning 9.1-12.0 microm wavelengths (lambda). Molecular vibration and vibration-rotation activities by the subject analyte are fundamentally strong within this "fingerprint" middle infrared spectral region. Distinct polarization-modulations of incident irradiance and backscatter radiance of tuned beams generate the Mueller matrix (M) of subject aerosol. Strings of all 15 normalized elements {M(ij)(lambda)/M(11)(lambda)}, which completely describe physical and geometric attributes of the aerosol particles, are input fields for training hybrid Kohonen self-organizing map feed-forward artificial neural networks (ANNs). The properly trained and validated ANN model performs pattern recognition and type-classification tasks via internal mappings. A typical ANN that mathematically clusters analyte, interferent, and control aerosols with nil overlap of species is illustrated, including sensitivity analysis of performance.

Relative optical navigation around small bodies via Extreme Learning Machine

NASA Astrophysics Data System (ADS)

Law, Andrew M.

To perform close proximity operations under a low-gravity environment, relative and absolute positions are vital information to the maneuver. Hence navigation is inseparably integrated in space travel. Extreme Learning Machine (ELM) is presented as an optical navigation method around small celestial bodies. Optical Navigation uses visual observation instruments such as a camera to acquire useful data and determine spacecraft position. The required input data for operation is merely a single image strip and a nadir image. ELM is a machine learning Single Layer feed-Forward Network (SLFN), a type of neural network (NN). The algorithm is developed on the predicate that input weights and biases can be randomly assigned and does not require back-propagation. The learned model is the output layer weights which are used to calculate a prediction. Together, Extreme Learning Machine Optical Navigation (ELM OpNav) utilizes optical images and ELM algorithm to train the machine to navigate around a target body. In this thesis the asteroid, Vesta, is the designated celestial body. The trained ELMs estimate the position of the spacecraft during operation with a single data set. The results show the approach is promising and potentially suitable for on-board navigation.

Generalized Adaptive Artificial Neural Networks

NASA Technical Reports Server (NTRS)

Tawel, Raoul

1993-01-01

Mathematical model of supervised learning by artificial neural network provides for simultaneous adjustments of both temperatures of neurons and synaptic weights, and includes feedback as well as feedforward synaptic connections. Extension of mathematical model described in "Adaptive Neurons For Artificial Neural Networks" (NPO-17803). Dynamics of neural network represented in new model by less-restrictive continuous formalism.

Neural Correlation Is Stimulus Modulated by Feedforward Inhibitory Circuitry

PubMed Central

Middleton, Jason W.; Omar, Cyrus; Doiron, Brent; Simons, Daniel J.

2012-01-01

Correlated variability of neural spiking activity has important consequences for signal processing. How incoming sensory signals shape correlations of population responses remains unclear. Cross-correlations between spiking of different neurons may be particularly consequential in sparsely firing neural populations such as those found in layer 2/3 of sensory cortex. In rat whisker barrel cortex, we found that pairs of excitatory layer 2/3 neurons exhibit similarly low levels of spike count correlation during both spontaneous and sensory-evoked states. The spontaneous activity of excitatory–inhibitory neuron pairs is positively correlated, while sensory stimuli actively decorrelate joint responses. Computational modeling shows how threshold nonlinearities and local inhibition form the basis of a general decorrelating mechanism. We show that inhibitory population activity maintains low correlations in excitatory populations, especially during periods of sensory-evoked coactivation. The role of feedforward inhibition has been previously described in the context of trial-averaged phenomena. Our findings reveal a novel role for inhibition to shape correlations of neural variability and thereby prevent excessive correlations in the face of feedforward sensory-evoked activation. PMID:22238086

A new hysteresis model based on force-displacement characteristics of magnetorheological fluid actuators subjected to squeeze mode operation

NASA Astrophysics Data System (ADS)

Chen, Peng; Bai, Xian-Xu; Qian, Li-Jun; Choi, Seung-Bok

2017-06-01

This paper presents a new hysteresis model based on the force-displacement characteristics of magnetorheological (MR) fluid actuators (or devices) subjected to squeeze mode operation. The idea of the proposed model is originated from experimental observation of the field-dependent hysteretic behavior of MR fluids, which shows that from a view of rate-independence of hysteresis, a gap width-dependent hysteresis is occurred in the force-displacement relationship instead of the typical relationship of the force-velocity. To effectively and accurately portray the hysteresis behavior, the gap width-dependent hysteresis elements, the nonlinear viscous effect and the inertial effect are considered for the formulation of the hysteresis model. Then, a model-based feedforward force tracking control scheme is established through an observer which can estimate the virtual displacement. The effectiveness of the proposed hysteresis model is validated through the identification and prediction of the damping force of MR fluids in the squeeze mode. In addition, it is shown that superior force tracking performance of the feedforward control associated with the proposed hysteresis mode is evaluated by adopting several tracking trajectories.

Kinetic model-based feed-forward controlled fed-batch fermentation of Lactobacillus rhamnosus for the production of lactic acid from Arabic date juice.

PubMed

Choi, Minsung; Al-Zahrani, Saeed M; Lee, Sang Yup

2014-06-01

Arabic date is overproduced in Arabic countries such as Saudi Arabia and Iraq and is mostly composed of sugars (70-80 wt%). Here we developed a fed-batch fermentation process by using a kinetic model for the efficient production of lactic acid to a high concentration from Arabic date juice. First, a kinetic model of Lactobacillus rhamnosus grown on date juice in batch fermentation was constructed in EXCEL so that the estimation of parameters and simulation of the model can be easily performed. Then, several fed-batch fermentations were conducted by employing different feeding strategies including pulsed feeding, exponential feeding, and modified exponential feeding. Based on the results of fed-batch fermentations, the kinetic model for fed-batch fermentation was also developed. This new model was used to perform feed-forward controlled fed-batch fermentation, which resulted in the production of 171.79 g l(-1) of lactic acid with the productivity and yield of 1.58 and 0.87 g l(-1) h(-1), respectively.

[Membrane-bound cytokine and feedforward regulation].

PubMed

Wu, Ke-Fu; Zheng, Guo-Guang; Ma, Xiao-Tong; Song, Yu-Hua

2013-10-01

Feedback and feedforward widely exist in life system, both of them are the basic processes of control system. While the concept of feedback has been widely used in life science, feedforward regulation was systematically studied in neurophysiology, awaiting further evidence and mechanism in molecular biology and cell biology. The authors put forward a hypothesis about the feedforward regulation of membrane bound macrophage colony stimulation factor (mM-CSF) on the basis of their previous work. This hypothesis might provide a new direction for the study on the biological effects of mM-CSF on leukemia and solid tumors, and contribute to the study on other membrane bound cytokines.

Support surface related changes in feedforward and feedback control of standing posture

PubMed Central

Mohapatra, Sambit; Kukkar, Komal K.; Aruin, Alexander S.

2013-01-01

The aim of the study was to investigate the effect of different support surfaces on feedforward and feedback components of postural control. Nine healthy subjects were exposed to external perturbations applied to their shoulders while standing on a rigid platform, foam, and wobble board with eyes open or closed. Electrical activity of nine trunk and leg muscles and displacements of the center of pressure were recorded and analyzed during the time frames typical of feedforward and feedback postural adjustments. Feedforward control of posture was characterized by earlier activation of anterior muscles when the subjects stood on foam compared to a wobble board or a firm surface. In addition, the magnitude of feedforward muscle activity was the largest when the foam was used. During the feedback control, anterior muscles were activated prior to posterior muscles irrespective of the nature of surface. Moreover, the largest muscle activity was seen when the supporting surface was foam. Maximum CoP displacement occurred when subjects were standing on a rigid surface. Altering support surface affects both feedforward and feedback components of postural control. This information should be taken into consideration in planning rehabilitation interventions geared towards improvement of balance. PMID:24268589

Support surface related changes in feedforward and feedback control of standing posture.

PubMed

Mohapatra, Sambit; Kukkar, Komal K; Aruin, Alexander S

2014-02-01

The aim of the study was to investigate the effect of different support surfaces on feedforward and feedback components of postural control. Nine healthy subjects were exposed to external perturbations applied to their shoulders while standing on a rigid platform, foam, and wobble board with eyes open or closed. Electrical activity of nine trunk and leg muscles and displacements of the center of pressure were recorded and analyzed during the time frames typical of feedforward and feedback postural adjustments. Feedforward control of posture was characterized by earlier activation of anterior muscles when the subjects stood on foam compared to a wobble board or a firm surface. In addition, the magnitude of feedforward muscle activity was the largest when the foam was used. During the feedback control, anterior muscles were activated prior to posterior muscles irrespective of the nature of surface. Moreover, the largest muscle activity was seen when the supporting surface was foam. Maximum CoP displacement occurred when subjects were standing on a rigid surface. Altering support surface affects both feedforward and feedback components of postural control. This information should be taken into consideration in planning rehabilitation interventions geared towards improvement of balance. Copyright © 2013 Elsevier Ltd. All rights reserved.

A comprehensive inversion approach for feedforward compensation of piezoactuator system at high frequency

NASA Astrophysics Data System (ADS)

Tian, Lizhi; Xiong, Zhenhua; Wu, Jianhua; Ding, Han

2016-09-01

Motion control of the piezoactuator system over broadband frequencies is limited due to its inherent hysteresis and system dynamics. One of the suggested ways is to use feedforward controller to linearize the input-output relationship of the piezoactuator system. Although there have been many feedforward approaches, it is still a challenge to develop feedforward controller for the piezoactuator system at high frequency. Hence, this paper presents a comprehensive inversion approach in consideration of the coupling of hysteresis and dynamics. In this work, the influence of dynamics compensation on the input-output relationship of the piezoactuator system is investigated first. With system dynamics compensation, the input-output relationship of the piezoactuator system will be further represented as rate-dependent nonlinearity due to the inevitable dynamics compensation error, especially at high frequency. Base on this result, the feedforward controller composed by a cascade of linear dynamics inversion and rate-dependent nonlinearity inversion is developed. Then, the system identification of the comprehensive inversion approach is proposed. Finally, experimental results show that the proposed approach can improve the performance on tracking of both periodic and non-periodic trajectories at medium and high frequency compared with the conventional feedforward approaches.

Analysis and Modeling of DIII-D Experiments With OMFIT and Neural Networks

NASA Astrophysics Data System (ADS)

Meneghini, O.; Luna, C.; Smith, S. P.; Lao, L. L.; GA Theory Team

2013-10-01

The OMFIT integrated modeling framework is designed to facilitate experimental data analysis and enable integrated simulations. This talk introduces this framework and presents a selection of its applications to the DIII-D experiment. Examples include kinetic equilibrium reconstruction analysis; evaluation of MHD stability in the core and in the edge; and self-consistent predictive steady-state transport modeling. The OMFIT framework also provides the platform for an innovative approach based on neural networks to predict electron and ion energy fluxes. In our study a multi-layer feed-forward back-propagation neural network is built and trained over a database of DIII-D data. It is found that given the same parameters that the highest fidelity models use, the neural network model is able to predict to a large degree the heat transport profiles observed in the DIII-D experiments. Once the network is built, the numerical cost of evaluating the transport coefficients is virtually nonexistent, thus making the neural network model particularly well suited for plasma control and quick exploration of operational scenarios. The implementation of the neural network model and benchmark with experimental results and gyro-kinetic models will be discussed. Work supported in part by the US DOE under DE-FG02-95ER54309.

Using wavelet-feedforward neural networks to improve air pollution forecasting in urban environments.

PubMed

Dunea, Daniel; Pohoata, Alin; Iordache, Stefania

2015-07-01

The paper presents the screening of various feedforward neural networks (FANN) and wavelet-feedforward neural networks (WFANN) applied to time series of ground-level ozone (O3), nitrogen dioxide (NO2), and particulate matter (PM10 and PM2.5 fractions) recorded at four monitoring stations located in various urban areas of Romania, to identify common configurations with optimal generalization performance. Two distinct model runs were performed as follows: data processing using hourly-recorded time series of airborne pollutants during cold months (O3, NO2, and PM10), when residential heating increases the local emissions, and data processing using 24-h daily averaged concentrations (PM2.5) recorded between 2009 and 2012. Dataset variability was assessed using statistical analysis. Time series were passed through various FANNs. Each time series was decomposed in four time-scale components using three-level wavelets, which have been passed also through FANN, and recomposed into a single time series. The agreement between observed and modelled output was evaluated based on the statistical significance (r coefficient and correlation between errors and data). Daubechies db3 wavelet-Rprop FANN (6-4-1) utilization gave positive results for O3 time series optimizing the exclusive use of the FANN for hourly-recorded time series. NO2 was difficult to model due to time series specificity, but wavelet integration improved FANN performances. Daubechies db3 wavelet did not improve the FANN outputs for PM10 time series. Both models (FANN/WFANN) overestimated PM2.5 forecasted values in the last quarter of time series. A potential improvement of the forecasted values could be the integration of a smoothing algorithm to adjust the PM2.5 model outputs.

Physics-model-based nonlinear actuator trajectory optimization and safety factor profile feedback control for advanced scenario development in DIII-D

DOE PAGES

Barton, Justin E.; Boyer, Mark D.; Shi, Wenyu; ...

2015-07-30

DIII-D experimental results are reported to demonstrate the potential of physics-model-based safety factor profile control for robust and reproducible sustainment of advanced scenarios. In the absence of feedback control, variability in wall conditions and plasma impurities, as well as drifts due to external disturbances, can limit the reproducibility of discharges with simple pre-programmed scenario trajectories. The control architecture utilized is a feedforward + feedback scheme where the feedforward commands are computed off-line and the feedback commands are computed on-line. In this work, firstly a first-principles-driven (FPD), physics-based model of the q profile and normalized beta (β N) dynamics is embeddedmore » into a numerical optimization algorithm to design feedforward actuator trajectories that sheer the plasma through the tokamak operating space to reach a desired stationary target state that is characterized by the achieved q profile and β N. Good agreement between experimental results and simulations demonstrates the accuracy of the models employed for physics-model-based control design. Secondly, a feedback algorithm for q profile control is designed following a FPD approach, and the ability of the controller to achieve and maintain a target q profile evolution is tested in DIII-D high confinement (H-mode) experiments. The controller is shown to be able to effectively control the q profile when β N is relatively close to the target, indicating the need for integrated q profile and β N control to further enhance the ability to achieve robust scenario execution. Furthermore, the ability of an integrated q profile + β N feedback controller to track a desired target is demonstrated through simulation.« less

A decentralized training algorithm for Echo State Networks in distributed big data applications.

PubMed

Scardapane, Simone; Wang, Dianhui; Panella, Massimo

2016-06-01

The current big data deluge requires innovative solutions for performing efficient inference on large, heterogeneous amounts of information. Apart from the known challenges deriving from high volume and velocity, real-world big data applications may impose additional technological constraints, including the need for a fully decentralized training architecture. While several alternatives exist for training feed-forward neural networks in such a distributed setting, less attention has been devoted to the case of decentralized training of recurrent neural networks (RNNs). In this paper, we propose such an algorithm for a class of RNNs known as Echo State Networks. The algorithm is based on the well-known Alternating Direction Method of Multipliers optimization procedure. It is formulated only in terms of local exchanges between neighboring agents, without reliance on a coordinating node. Additionally, it does not require the communication of training patterns, which is a crucial component in realistic big data implementations. Experimental results on large scale artificial datasets show that it compares favorably with a fully centralized implementation, in terms of speed, efficiency and generalization accuracy. Copyright © 2015 Elsevier Ltd. All rights reserved.

Variable synaptic strengths controls the firing rate distribution in feedforward neural networks.

PubMed

Ly, Cheng; Marsat, Gary

2018-02-01

Heterogeneity of firing rate statistics is known to have severe consequences on neural coding. Recent experimental recordings in weakly electric fish indicate that the distribution-width of superficial pyramidal cell firing rates (trial- and time-averaged) in the electrosensory lateral line lobe (ELL) depends on the stimulus, and also that network inputs can mediate changes in the firing rate distribution across the population. We previously developed theoretical methods to understand how two attributes (synaptic and intrinsic heterogeneity) interact and alter the firing rate distribution in a population of integrate-and-fire neurons with random recurrent coupling. Inspired by our experimental data, we extend these theoretical results to a delayed feedforward spiking network that qualitatively capture the changes of firing rate heterogeneity observed in in-vivo recordings. We demonstrate how heterogeneous neural attributes alter firing rate heterogeneity, accounting for the effect with various sensory stimuli. The model predicts how the strength of the effective network connectivity is related to intrinsic heterogeneity in such delayed feedforward networks: the strength of the feedforward input is positively correlated with excitability (threshold value for spiking) when firing rate heterogeneity is low and is negatively correlated with excitability with high firing rate heterogeneity. We also show how our theory can be used to predict effective neural architecture. We demonstrate that neural attributes do not interact in a simple manner but rather in a complex stimulus-dependent fashion to control neural heterogeneity and discuss how it can ultimately shape population codes.

Startle stimuli reduce the internal model control in discrete movements.

PubMed

Wright, Zachary A; Rogers, Mark W; MacKinnon, Colum D; Patton, James L

2009-01-01

A well known and major component of movement control is the feedforward component, also known as the internal model. This model predicts and compensates for expected forces seen during a movement, based on recent experience, so that a well-learned task such as reaching to a target can be executed in a smooth straight manner. It has recently been shown that the state of preparation of planned movements can be tested using a startling acoustic stimulus (SAS). SAS, presented 500, 250 or 0 ms before the expected "go" cue resulted in the early release of the movement trajectory associated with the after-effects of the force field training (i.e. the internal model). In a typical motor adaptation experiment with a robot-applied force field, we tested if a SAS stimulus influences the size of after-effects that are typically seen. We found that in all subjects the after-effect magnitudes were significantly reduced when movements were released by SAS, although this effect was not further modulated by the timing of SAS. Reduced after-effects reveal at least partial existence of learned preparatory control, and identify startle effects that could influence performance in tasks such as piloting, teleoperation, and sports.

Modeling constitutive behavior of a 15Cr-15Ni-2.2Mo-Ti modified austenitic stainless steel under hot compression using artificial neural network

NASA Astrophysics Data System (ADS)

Mandal, Sumantra

2006-11-01

ABSTRACT In this paper, an artificial neural network (ANN) model has been suggested to predict the constitutive flow behavior of a 15Cr-15Ni-2.2Mo-Ti modified austenitic stainless steel under hot deformation. Hot compression tests in the temperature range 850°C- 1250°C and strain rate range 10-3-102 s-1 were carried out. These tests provided the required data for training the neural network and for subsequent testing. The inputs of the neural network are strain, log strain rate and temperature while flow stress is obtained as output. A three layer feed-forward network with ten neurons in a single hidden layer and back-propagation learning algorithm has been employed. A very good correlation between experimental and predicted result has been obtained. The effect of temperature and strain rate on flow behavior has been simulated employing the ANN model. The results have been found to be consistent with the metallurgical trend. Finally, a monte carlo analiysis has been carried out to find out the noise sensitivity of the developed model.

Applications of Artificial Neural Networks in Structural Engineering with Emphasis on Continuum Models

NASA Technical Reports Server (NTRS)

Kapania, Rakesh K.; Liu, Youhua

1998-01-01

The use of continuum models for the analysis of discrete built-up complex aerospace structures is an attractive idea especially at the conceptual and preliminary design stages. But the diversity of available continuum models and hard-to-use qualities of these models have prevented them from finding wide applications. In this regard, Artificial Neural Networks (ANN or NN) may have a great potential as these networks are universal approximators that can realize any continuous mapping, and can provide general mechanisms for building models from data whose input-output relationship can be highly nonlinear. The ultimate aim of the present work is to be able to build high fidelity continuum models for complex aerospace structures using the ANN. As a first step, the concepts and features of ANN are familiarized through the MATLAB NN Toolbox by simulating some representative mapping examples, including some problems in structural engineering. Then some further aspects and lessons learned about the NN training are discussed, including the performances of Feed-Forward and Radial Basis Function NN when dealing with noise-polluted data and the technique of cross-validation. Finally, as an example of using NN in continuum models, a lattice structure with repeating cells is represented by a continuum beam whose properties are provided by neural networks.

A self-cognizant dynamic system approach for prognostics and health management

NASA Astrophysics Data System (ADS)

Bai, Guangxing; Wang, Pingfeng; Hu, Chao

2015-03-01

Prognostics and health management (PHM) is an emerging engineering discipline that diagnoses and predicts how and when a system will degrade its performance and lose its partial or whole functionality. Due to the complexity and invisibility of rules and states of most dynamic systems, developing an effective approach to track evolving system states becomes a major challenge. This paper presents a new self-cognizant dynamic system (SCDS) approach that incorporates artificial intelligence into dynamic system modeling for PHM. A feed-forward neural network (FFNN) is selected to approximate a complex system response which is challenging task in general due to inaccessible system physics. The trained FFNN model is then embedded into a dual extended Kalman filter algorithm to track down system dynamics. A recursive computation technique used to update the FFNN model using online measurements is also derived. To validate the proposed SCDS approach, a battery dynamic system is considered as an experimental application. After modeling the battery system by a FFNN model and a state-space model, the state-of-charge (SoC) and state-of-health (SoH) are estimated by updating the FFNN model using the proposed approach. Experimental results suggest that the proposed approach improves the efficiency and accuracy for battery health management.

Feed-Forward Neural Network Prediction of the Mechanical Properties of Sandcrete Materials

PubMed Central

Asteris, Panagiotis G.; Roussis, Panayiotis C.; Douvika, Maria G.

2017-01-01

This work presents a soft-sensor approach for estimating critical mechanical properties of sandcrete materials. Feed-forward (FF) artificial neural network (ANN) models are employed for building soft-sensors able to predict the 28-day compressive strength and the modulus of elasticity of sandcrete materials. To this end, a new normalization technique for the pre-processing of data is proposed. The comparison of the derived results with the available experimental data demonstrates the capability of FF ANNs to predict with pinpoint accuracy the mechanical properties of sandcrete materials. Furthermore, the proposed normalization technique has been proven effective and robust compared to other normalization techniques available in the literature. PMID:28598400

Artificial neural networks in Space Station optimal attitude control

NASA Astrophysics Data System (ADS)

Kumar, Renjith R.; Seywald, Hans; Deshpande, Samir M.; Rahman, Zia

1995-01-01

Innovative techniques of using "artificial neural networks" (ANN) for improving the performance of the pitch axis attitude control system of Space Station Freedom using control moment gyros (CMGs) are investigated. The first technique uses a feed-forward ANN with multi-layer perceptrons to obtain an on-line controller which improves the performance of the control system via a model following approach. The second technique uses a single layer feed-forward ANN with a modified back propagation scheme to estimate the internal plant variations and the external disturbances separately. These estimates are then used to solve two differential Riccati equations to obtain time varying gains which improve the control system performance in successive orbits.

Impaired Feedforward Control and Enhanced Feedback Control of Speech in Patients with Cerebellar Degeneration.

PubMed

Parrell, Benjamin; Agnew, Zarinah; Nagarajan, Srikantan; Houde, John; Ivry, Richard B

2017-09-20

The cerebellum has been hypothesized to form a crucial part of the speech motor control network. Evidence for this comes from patients with cerebellar damage, who exhibit a variety of speech deficits, as well as imaging studies showing cerebellar activation during speech production in healthy individuals. To date, the precise role of the cerebellum in speech motor control remains unclear, as it has been implicated in both anticipatory (feedforward) and reactive (feedback) control. Here, we assess both anticipatory and reactive aspects of speech motor control, comparing the performance of patients with cerebellar degeneration and matched controls. Experiment 1 tested feedforward control by examining speech adaptation across trials in response to a consistent perturbation of auditory feedback. Experiment 2 tested feedback control, examining online corrections in response to inconsistent perturbations of auditory feedback. Both male and female patients and controls were tested. The patients were impaired in adapting their feedforward control system relative to controls, exhibiting an attenuated anticipatory response to the perturbation. In contrast, the patients produced even larger compensatory responses than controls, suggesting an increased reliance on sensory feedback to guide speech articulation in this population. Together, these results suggest that the cerebellum is crucial for maintaining accurate feedforward control of speech, but relatively uninvolved in feedback control. SIGNIFICANCE STATEMENT Speech motor control is a complex activity that is thought to rely on both predictive, feedforward control as well as reactive, feedback control. While the cerebellum has been shown to be part of the speech motor control network, its functional contribution to feedback and feedforward control remains controversial. Here, we use real-time auditory perturbations of speech to show that patients with cerebellar degeneration are impaired in adapting feedforward control of speech but retain the ability to make online feedback corrections; indeed, the patients show an increased sensitivity to feedback. These results indicate that the cerebellum forms a crucial part of the feedforward control system for speech but is not essential for online, feedback control. Copyright © 2017 the authors 0270-6474/17/379249-10$15.00/0.

Adaptive Control Of Remote Manipulator

NASA Technical Reports Server (NTRS)

Seraji, Homayoun

1989-01-01

Robotic control system causes remote manipulator to follow closely reference trajectory in Cartesian reference frame in work space, without resort to computationally intensive mathematical model of robot dynamics and without knowledge of robot and load parameters. System, derived from linear multivariable theory, uses relatively simple feedforward and feedback controllers with model-reference adaptive control.

On the sample complexity of learning for networks of spiking neurons with nonlinear synaptic interactions.

PubMed

Schmitt, Michael

2004-09-01

We study networks of spiking neurons that use the timing of pulses to encode information. Nonlinear interactions model the spatial groupings of synapses on the neural dendrites and describe the computations performed at local branches. Within a theoretical framework of learning we analyze the question of how many training examples these networks must receive to be able to generalize well. Bounds for this sample complexity of learning can be obtained in terms of a combinatorial parameter known as the pseudodimension. This dimension characterizes the computational richness of a neural network and is given in terms of the number of network parameters. Two types of feedforward architectures are considered: constant-depth networks and networks of unconstrained depth. We derive asymptotically tight bounds for each of these network types. Constant depth networks are shown to have an almost linear pseudodimension, whereas the pseudodimension of general networks is quadratic. Networks of spiking neurons that use temporal coding are becoming increasingly more important in practical tasks such as computer vision, speech recognition, and motor control. The question of how well these networks generalize from a given set of training examples is a central issue for their successful application as adaptive systems. The results show that, although coding and computation in these networks is quite different and in many cases more powerful, their generalization capabilities are at least as good as those of traditional neural network models.

Mexican Hat Wavelet Kernel ELM for Multiclass Classification.

PubMed

Wang, Jie; Song, Yi-Fan; Ma, Tian-Lei

2017-01-01

Kernel extreme learning machine (KELM) is a novel feedforward neural network, which is widely used in classification problems. To some extent, it solves the existing problems of the invalid nodes and the large computational complexity in ELM. However, the traditional KELM classifier usually has a low test accuracy when it faces multiclass classification problems. In order to solve the above problem, a new classifier, Mexican Hat wavelet KELM classifier, is proposed in this paper. The proposed classifier successfully improves the training accuracy and reduces the training time in the multiclass classification problems. Moreover, the validity of the Mexican Hat wavelet as a kernel function of ELM is rigorously proved. Experimental results on different data sets show that the performance of the proposed classifier is significantly superior to the compared classifiers.

Synthesis of recurrent neural networks for dynamical system simulation.

PubMed

Trischler, Adam P; D'Eleuterio, Gabriele M T

2016-08-01

We review several of the most widely used techniques for training recurrent neural networks to approximate dynamical systems, then describe a novel algorithm for this task. The algorithm is based on an earlier theoretical result that guarantees the quality of the network approximation. We show that a feedforward neural network can be trained on the vector-field representation of a given dynamical system using backpropagation, then recast it as a recurrent network that replicates the original system's dynamics. After detailing this algorithm and its relation to earlier approaches, we present numerical examples that demonstrate its capabilities. One of the distinguishing features of our approach is that both the original dynamical systems and the recurrent networks that simulate them operate in continuous time. Copyright © 2016 Elsevier Ltd. All rights reserved.

Designing an Adaptive Web-Based Learning System Based on Students' Cognitive Styles Identified Online

ERIC Educational Resources Information Center

Lo, Jia-Jiunn; Chan, Ya-Chen; Yeh, Shiou-Wen

2012-01-01

This study developed an adaptive web-based learning system focusing on students' cognitive styles. The system is composed of a student model and an adaptation model. It collected students' browsing behaviors to update the student model for unobtrusively identifying student cognitive styles through a multi-layer feed-forward neural network (MLFF).…

Averaging of phase noise in PSK signals by an opto-electrical feed-forward circuit

NASA Astrophysics Data System (ADS)

Inoue, K.; Ohta, M.

2013-10-01

This paper proposes an opto-electrical feed-forward circuit that reduces phase noise in binary PSK signals by averaging the noise. Random and independent phase noise is averaged over several bit slots by externally modulating a phase-fluctuating PSK signal with feed-forward signal obtained from signal processing of the outputs of delay interferometers. The simulation results demonstrate a reduction in the phase noise.

Long-Term Adaptations to Unexpected Surface Perturbations: Postural Control During Stance and Gait in Train Conductors.

PubMed

Baumgart, Christian; Hoppe, Matthias Wilhelm; Freiwald, Jürgen

2016-01-01

The authors aimed to evaluate the differences in postural control during stance and gait between train conductors and controls. Twenty-one train conductors and 21 office workers performed 6 unilateral and bilateral balance tests on stable and unstable surfaces as well as a gait analysis. In the balance tests, the mean velocity of the center of pressure and unstable surface was measured. In the bilateral balance tests the selected stance width was measured. During gait the length, width, frequency, and velocity of the steps were calculated from the ground reaction forces. Train conductors showed a significantly greater step width during gait (15.4 ± 4.7 vs. 13.0 ± 3.4 cm; p = .035) and stance width during the bilateral stance on the unstable surface (21.0 ± 5.1 vs. 17.8 ± 3.7 cm; p = .026) than the office workers, while no differences were revealed in balance variables. The revealed differences between train conductors and office workers may represent task-specific feedforward control strategies, which increase the base of support and may be helpful to resist unexpected perturbations in trains.

Stability and Control of Human Trunk Movement During Walking.

PubMed

Wu, Q.; Sepehri, N.; Thornton-Trump, A. B.; Alexander, M.

1998-01-01

A mathematical model has been developed to study the control mechanisms of human trunk movement during walking. The trunk is modeled as a base-excited inverted pendulum with two-degrees of rotational freedom. The base point, corresponding to the bony landmark of the sacrum, can move in three-dimensional space in a general way. Since the stability of upright posture is essential for human walking, a controller has been designed such that the stability of the pendulum about the upright position is guaranteed. The control laws are developed based on Lyapunov's stability theory and include feedforward and linear feedback components. It is found that the feedforward component plays a critical role in keeping postural stability, and the linear feedback component, (resulting from viscoelastic function of the musculoskeletal system) can effectively duplicate the pattern of trunk movement. The mathematical model is validated by comparing the simulation results with those based on gait measurements performed in the Biomechanics Laboratory at the University of Manitoba.

Proinflammatory Cytokine Infusion Attenuates LH's Feedforward on Testosterone Secretion: Modulation by Age

PubMed Central

Yang, Rebecca; Roelfsema, Ferdinand; Takahashi, Paul

2016-01-01

Context: In the experimental animal, inflammatory signals quench LH's feedforward drive of testosterone (T) secretion and appear to impair GnRH-LH output. The degree to which such suppressive effects operate in the human is not known. Objective: To test the hypothesis that IL-2 impairs LH's feedforward drive on T and T's feedback inhibition of LH secretion in healthy men. Setting: Mayo Center for Translational Science Activities. Patients or Other Participants: A total of 35 healthy men, 17 young and 18 older. Interventions: Randomized prospective double-blind saline-controlled study of IL-2 infusion in 2 doses with concurrent 10-minute blood sampling for 24 hours. Main Outcome Measures: Deconvolution analysis of LH and T secretion. Results: After saline injection, older compared with young men exhibited reduced LH feedforward drive on T secretion (P < .001), and decreased T feedback inhibition of LH secretion (P < .01). After IL-2 injection, LH's feedforward onto T secretion declined markedly especially in young subjects (P < .001). Concomitantly, IL-2 potentiated T's proportional feedback on LH secretion especially in older volunteers. Conclusion: This investigation confirms combined feedforward and feedback deficits in older relative to young men given saline and demonstrates 1) joint mechanisms by which IL-2 enforces biochemical hypogonadism, viz, combined feedforward block and feedback amplification; and 2) unequal absolute inhibition of T and LH secretion by IL-2 in young and older men. These outcomes establish that the male gonadal axis is susceptible to dual-site suppression by a prototypic inflammatory mediator. Thus, we postulate that selected ILs might also enforce male hypogonadism in chronic systemic inflammation. PMID:26600270

Proinflammatory Cytokine Infusion Attenuates LH's Feedforward on Testosterone Secretion: Modulation by Age.

PubMed

Veldhuis, Johannes; Yang, Rebecca; Roelfsema, Ferdinand; Takahashi, Paul

2016-02-01

In the experimental animal, inflammatory signals quench LH's feedforward drive of testosterone (T) secretion and appear to impair GnRH-LH output. The degree to which such suppressive effects operate in the human is not known. To test the hypothesis that IL-2 impairs LH's feedforward drive on T and T's feedback inhibition of LH secretion in healthy men. Mayo Center for Translational Science Activities. A total of 35 healthy men, 17 young and 18 older. Randomized prospective double-blind saline-controlled study of IL-2 infusion in 2 doses with concurrent 10-minute blood sampling for 24 hours. Deconvolution analysis of LH and T secretion. After saline injection, older compared with young men exhibited reduced LH feedforward drive on T secretion (P < .001), and decreased T feedback inhibition of LH secretion (P < .01). After IL-2 injection, LH's feedforward onto T secretion declined markedly especially in young subjects (P < .001). Concomitantly, IL-2 potentiated T's proportional feedback on LH secretion especially in older volunteers. This investigation confirms combined feedforward and feedback deficits in older relative to young men given saline and demonstrates 1) joint mechanisms by which IL-2 enforces biochemical hypogonadism, viz, combined feedforward block and feedback amplification; and 2) unequal absolute inhibition of T and LH secretion by IL-2 in young and older men. These outcomes establish that the male gonadal axis is susceptible to dual-site suppression by a prototypic inflammatory mediator. Thus, we postulate that selected ILs might also enforce male hypogonadism in chronic systemic inflammation.

Neural Correlates of Early Sound Encoding and their Relationship to Speech-in-Noise Perception

PubMed Central

Coffey, Emily B. J.; Chepesiuk, Alexander M. P.; Herholz, Sibylle C.; Baillet, Sylvain; Zatorre, Robert J.

2017-01-01

Speech-in-noise (SIN) perception is a complex cognitive skill that affects social, vocational, and educational activities. Poor SIN ability particularly affects young and elderly populations, yet varies considerably even among healthy young adults with normal hearing. Although SIN skills are known to be influenced by top-down processes that can selectively enhance lower-level sound representations, the complementary role of feed-forward mechanisms and their relationship to musical training is poorly understood. Using a paradigm that minimizes the main top-down factors that have been implicated in SIN performance such as working memory, we aimed to better understand how robust encoding of periodicity in the auditory system (as measured by the frequency-following response) contributes to SIN perception. Using magnetoencephalograpy, we found that the strength of encoding at the fundamental frequency in the brainstem, thalamus, and cortex is correlated with SIN accuracy. The amplitude of the slower cortical P2 wave was previously also shown to be related to SIN accuracy and FFR strength; we use MEG source localization to show that the P2 wave originates in a temporal region anterior to that of the cortical FFR. We also confirm that the observed enhancements were related to the extent and timing of musicianship. These results are consistent with the hypothesis that basic feed-forward sound encoding affects SIN perception by providing better information to later processing stages, and that modifying this process may be one mechanism through which musical training might enhance the auditory networks that subserve both musical and language functions. PMID:28890684

The decay of motor adaptation to novel movement dynamics reveals an asymmetry in the stability of motion state-dependent learning

PubMed Central

Hosseini, Eghbal A.; Nguyen, Katrina P.; Joiner, Wilsaan M.

2017-01-01

Motor adaptation paradigms provide a quantitative method to study short-term modification of motor commands. Despite the growing understanding of the role motion states (e.g., velocity) play in this form of motor learning, there is little information on the relative stability of memories based on these movement characteristics, especially in comparison to the initial adaptation. Here, we trained subjects to make reaching movements perturbed by force patterns dependent upon either limb position or velocity. Following training, subjects were exposed to a series of error-clamp trials to measure the temporal characteristics of the feedforward motor output during the decay of learning. The compensatory force patterns were largely based on the perturbation kinematic (e.g., velocity), but also showed a small contribution from the other motion kinematic (e.g., position). However, the velocity contribution in response to the position-based perturbation decayed at a slower rate than the position contribution to velocity-based training, suggesting a difference in stability. Next, we modified a previous model of motor adaptation to reflect this difference and simulated the behavior for different learning goals. We were interested in the stability of learning when the perturbations were based on different combinations of limb position or velocity that subsequently resulted in biased amounts of motion-based learning. We trained additional subjects on these combined motion-state perturbations and confirmed the predictions of the model. Specifically, we show that (1) there is a significant separation between the observed gain-space trajectories for the learning and decay of adaptation and (2) for combined motion-state perturbations, the gain associated to changes in limb position decayed at a faster rate than the velocity-dependent gain, even when the position-dependent gain at the end of training was significantly greater. Collectively, these results suggest that the state-dependent adaptation associated with movement velocity is relatively more stable than that based on position. PMID:28481891

Feedback Enhances Feedforward Figure-Ground Segmentation by Changing Firing Mode

PubMed Central

Supèr, Hans; Romeo, August

2011-01-01

In the visual cortex, feedback projections are conjectured to be crucial in figure-ground segregation. However, the precise function of feedback herein is unclear. Here we tested a hypothetical model of reentrant feedback. We used a previous developed 2-layered feedforwardspiking network that is able to segregate figure from ground and included feedback connections. Our computer model data show that without feedback, neurons respond with regular low-frequency (∼9 Hz) bursting to a figure-ground stimulus. After including feedback the firing pattern changed into a regular (tonic) spiking pattern. In this state, we found an extra enhancement of figure responses and a further suppression of background responses resulting in a stronger figure-ground signal. Such push-pull effect was confirmed by comparing the figure-ground responses withthe responses to a homogenous texture. We propose that feedback controlsfigure-ground segregation by influencing the neural firing patterns of feedforward projecting neurons. PMID:21738747

Feedforward Self-Modeling Enhances Skill Acquisition in Children Learning Trampoline Skills

PubMed Central

Ste-Marie, Diane M.; Vertes, Kelly; Rymal, Amanda M.; Martini, Rose

2011-01-01

The purpose of this research was to examine whether children would benefit from a feedforward self-modeling (FSM) video and to explore possible explanatory mechanisms for the potential benefits, using a self-regulation framework. To this end, children were involved in learning two five-skill trampoline routines. For one of the routines, a FSM video was provided during acquisition, whereas only verbal instructions were provided for the alternate routine. The FSM involved editing video footage such that it showed the learner performing the trampoline routine at a higher skill level than their current capability. Analyses of the data showed that while physical performance benefits were observed for the routine that was learned with the FSM video, no differences were obtained in relation to the self-regulatory measures. Thus, the FSM video enhanced motor skill acquisition, but this could not be explained by changes to the varied self-regulatory processes examined. PMID:21779270

Control of a flexible bracing manipulator: Integration of current research work to realize the bracing manipulator

NASA Technical Reports Server (NTRS)

Kwon, Dong-Soo

1991-01-01

All research results about flexible manipulator control were integrated to show a control scenario of a bracing manipulator. First, dynamic analysis of a flexible manipulator was done for modeling. Second, from the dynamic model, the inverse dynamic equation was derived, and the time domain inverse dynamic method was proposed for the calculation of the feedforward torque and the desired flexible coordinate trajectories. Third, a tracking controller was designed by combining the inverse dynamic feedforward control with the joint feedback control. The control scheme was applied to the tip position control of a single link flexible manipulator for zero and non-zero initial condition cases. Finally, the contact control scheme was added to the position tracking control. A control scenario of a bracing manipulator is provided and evaluated through simulation and experiment on a single link flexible manipulator.

Adaptive control of robotic manipulators

NASA Technical Reports Server (NTRS)

Seraji, H.

1987-01-01

The author presents a novel approach to adaptive control of manipulators to achieve trajectory tracking by the joint angles. The central concept in this approach is the utilization of the manipulator inverse as a feedforward controller. The desired trajectory is applied as an input to the feedforward controller which behaves as the inverse of the manipulator at any operating point; the controller output is used as the driving torque for the manipulator. The controller gains are then updated by an adaptation algorithm derived from MRAC (model reference adaptive control) theory to cope with variations in the manipulator inverse due to changes of the operating point. An adaptive feedback controller and an auxiliary signal are also used to enhance closed-loop stability and to achieve faster adaptation. The proposed control scheme is computationally fast and does not require a priori knowledge of the complex dynamic model or the parameter values of the manipulator or the payload.

Feedforward self-modeling enhances skill acquisition in children learning trampoline skills.

PubMed

Ste-Marie, Diane M; Vertes, Kelly; Rymal, Amanda M; Martini, Rose

2011-01-01

The purpose of this research was to examine whether children would benefit from a feedforward self-modeling (FSM) video and to explore possible explanatory mechanisms for the potential benefits, using a self-regulation framework. To this end, children were involved in learning two five-skill trampoline routines. For one of the routines, a FSM video was provided during acquisition, whereas only verbal instructions were provided for the alternate routine. The FSM involved editing video footage such that it showed the learner performing the trampoline routine at a higher skill level than their current capability. Analyses of the data showed that while physical performance benefits were observed for the routine that was learned with the FSM video, no differences were obtained in relation to the self-regulatory measures. Thus, the FSM video enhanced motor skill acquisition, but this could not be explained by changes to the varied self-regulatory processes examined.

Temporal integration and 1/f power scaling in a circuit model of cerebellar interneurons.

PubMed

Maex, Reinoud; Gutkin, Boris

2017-07-01

Inhibitory interneurons interconnected via electrical and chemical (GABA A receptor) synapses form extensive circuits in several brain regions. They are thought to be involved in timing and synchronization through fast feedforward control of principal neurons. Theoretical studies have shown, however, that whereas self-inhibition does indeed reduce response duration, lateral inhibition, in contrast, may generate slow response components through a process of gradual disinhibition. Here we simulated a circuit of interneurons (stellate and basket cells) of the molecular layer of the cerebellar cortex and observed circuit time constants that could rise, depending on parameter values, to >1 s. The integration time scaled both with the strength of inhibition, vanishing completely when inhibition was blocked, and with the average connection distance, which determined the balance between lateral and self-inhibition. Electrical synapses could further enhance the integration time by limiting heterogeneity among the interneurons and by introducing a slow capacitive current. The model can explain several observations, such as the slow time course of OFF-beam inhibition, the phase lag of interneurons during vestibular rotation, or the phase lead of Purkinje cells. Interestingly, the interneuron spike trains displayed power that scaled approximately as 1/ f at low frequencies. In conclusion, stellate and basket cells in cerebellar cortex, and interneuron circuits in general, may not only provide fast inhibition to principal cells but also act as temporal integrators that build a very short-term memory. NEW & NOTEWORTHY The most common function attributed to inhibitory interneurons is feedforward control of principal neurons. In many brain regions, however, the interneurons are densely interconnected via both chemical and electrical synapses but the function of this coupling is largely unknown. Based on large-scale simulations of an interneuron circuit of cerebellar cortex, we propose that this coupling enhances the integration time constant, and hence the memory trace, of the circuit. Copyright © 2017 the American Physiological Society.

Prediction of Air Pollutants Concentration Based on an Extreme Learning Machine: The Case of Hong Kong

PubMed Central

Zhang, Jiangshe; Ding, Weifu

2017-01-01

With the development of the economy and society all over the world, most metropolitan cities are experiencing elevated concentrations of ground-level air pollutants. It is urgent to predict and evaluate the concentration of air pollutants for some local environmental or health agencies. Feed-forward artificial neural networks have been widely used in the prediction of air pollutants concentration. However, there are some drawbacks, such as the low convergence rate and the local minimum. The extreme learning machine for single hidden layer feed-forward neural networks tends to provide good generalization performance at an extremely fast learning speed. The major sources of air pollutants in Hong Kong are mobile, stationary, and from trans-boundary sources. We propose predicting the concentration of air pollutants by the use of trained extreme learning machines based on the data obtained from eight air quality parameters in two monitoring stations, including Sham Shui Po and Tap Mun in Hong Kong for six years. The experimental results show that our proposed algorithm performs better on the Hong Kong data both quantitatively and qualitatively. Particularly, our algorithm shows better predictive ability, with R2 increased and root mean square error values decreased respectively. PMID:28125034

Estimating surface longwave radiative fluxes from satellites utilizing artificial neural networks

NASA Astrophysics Data System (ADS)

Nussbaumer, Eric A.; Pinker, Rachel T.

2012-04-01

A novel approach for calculating downwelling surface longwave (DSLW) radiation under all sky conditions is presented. The DSLW model (hereafter, DSLW/UMD v2) similarly to its predecessor, DSLW/UMD v1, is driven with a combination of Moderate Resolution Imaging Spectroradiometer (MODIS) level-3 cloud parameters and information from the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA-Interim model. To compute the clear sky component of DSLW a two layer feed-forward artificial neural network with sigmoid hidden neurons and linear output neurons is implemented; it is trained with simulations derived from runs of the Rapid Radiative Transfer Model (RRTM). When computing the cloud contribution to DSLW, the cloud base temperature is estimated by using an independent artificial neural network approach of similar architecture as previously mentioned, and parameterizations. The cloud base temperature neural network is trained using spatially and temporally co-located MODIS and CloudSat Cloud Profiling Radar (CPR) and the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) observations. Daily average estimates of DSLW from 2003 to 2009 are compared against ground measurements from the Baseline Surface Radiation Network (BSRN) giving an overall correlation coefficient of 0.98, root mean square error (rmse) of 15.84 W m-2, and a bias of -0.39 W m-2. This is an improvement over an earlier version of the model (DSLW/UMD v1) which for the same time period has an overall correlation coefficient 0.97 rmse of 17.27 W m-2, and bias of 0.73 W m-2.

Model predictive control-based dynamic coordinate strategy for hydraulic hub-motor auxiliary system of a heavy commercial vehicle

NASA Astrophysics Data System (ADS)

Zeng, Xiaohua; Li, Guanghan; Yin, Guodong; Song, Dafeng; Li, Sheng; Yang, Nannan

2018-02-01

Equipping a hydraulic hub-motor auxiliary system (HHMAS), which mainly consists of a hydraulic variable pump, a hydraulic hub-motor, a hydraulic valve block and hydraulic accumulators, with part-time all-wheel-drive functions improves the power performance and fuel economy of heavy commercial vehicles. The coordinated control problem that occurs when HHMAS operates in the auxiliary drive mode is addressed in this paper; the solution to this problem is the key to the maximization of HHMAS. To achieve a reasonable distribution of the engine power between mechanical and hydraulic paths, a nonlinear control scheme based on model predictive control (MPC) is investigated. First, a nonlinear model of HHMAS with vehicle dynamics and tire slip characteristics is built, and a controller-design-oriented model is simplified. Then, a steady-state feedforward + dynamic MPC feedback controller (FMPC) is designed to calculate the control input sequence of engine torque and hydraulic variable pump displacement. Finally, the controller is tested in the MATLAB/Simulink and AMESim co-simulation platform and the hardware-in-the-loop experiment platform, and its performance is compared with that of the existing proportional-integral-derivative controller and the feedforward controller under the same conditions. Simulation results show that the designed FMPC has the best performance, and control performance can be guaranteed in a real-time environment. Compared with the tracking control error of the feedforward controller, that of the designed FMPC is decreased by 85% and the traction efficiency performance is improved by 23% under a low-friction-surface condition. Moreover, under common road conditions for heavy commercial vehicles, the traction force can increase up to 13.4-15.6%.

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

Quality by design for herbal drugs: a feedforward control strategy and an approach to define the acceptable ranges of critical quality attributes.

PubMed

Yan, Binjun; Li, Yao; Guo, Zhengtai; Qu, Haibin

2014-01-01

The concept of quality by design (QbD) has been widely accepted and applied in the pharmaceutical manufacturing industry. There are still two key issues to be addressed in the implementation of QbD for herbal drugs. The first issue is the quality variation of herbal raw materials and the second issue is the difficulty in defining the acceptable ranges of critical quality attributes (CQAs). To propose a feedforward control strategy and a method for defining the acceptable ranges of CQAs for the two issues. In the case study of the ethanol precipitation process of Danshen (Radix Salvia miltiorrhiza) injection, regression models linking input material attributes and process parameters to CQAs were built first and an optimisation model for calculating the best process parameters according to the input materials was established. Then, the feasible material space was defined and the acceptable ranges of CQAs for the previous process were determined. In the case study, satisfactory regression models were built with cross-validated regression coefficients (Q(2) ) all above 91 %. The feedforward control strategy was applied successfully to compensate the quality variation of the input materials, which was able to control the CQAs in the 90-110 % ranges of the desired values. In addition, the feasible material space for the ethanol precipitation process was built successfully, which showed the acceptable ranges of the CQAs for the concentration process. The proposed methodology can help to promote the implementation of QbD for herbal drugs. Copyright © 2013 John Wiley & Sons, Ltd.

Deficits of anticipatory grip force control after damage to peripheral and central sensorimotor systems.

PubMed

Hermsdörfer, Joachim; Hagl, Elke; Nowak, Dennis A

2004-11-01

Healthy subjects adjust their grip force economically to the weight of a hand-held object. In addition, inertial loads, which arise from arm movements with the grasped object, are anticipated by parallel grip force modulations. Internal forward models have been proposed to predict the consequences of voluntary movements. Anesthesia of the fingers impairs grip force economy but the feedforward character of the grip force/load coupling is preserved. To further analyze the role of sensory input for internal forward models and to characterize the consequences of central nervous system damage for anticipatory grip force control, we measured grip force behavior in neurological patients. We tested a group of stroke patients with varying degrees of impaired fine motor control and sensory loss, a single patient with complete and permanent differentation from all tactile and proprioceptive input, and a group of patients with amyotrophic lateral sclerosis (ALS) that exclusively impairs the motor system without affecting sensory modalities. Increased grip forces were a common finding in all patients. Sensory deficits were a strong but not the only predictor of impaired grip force economy. The feedforward mode of grip force control was typically preserved in the stroke patients despite their central sensory deficits, but was severely disturbed in the patient with peripheral sensory deafferentation and in a minority of stroke patients. Moderate deficits of feedforward control were also obvious in ALS patients. Thus, the function of the internal forward model and the precision of grip force production may depend on a complex anatomical and functional network of sensory and motor structures and their interaction in time and space.

Dynamic balance of excitation and inhibition rapidly modulates spike probability and precision in feed-forward hippocampal circuits

PubMed Central

Wahlstrom-Helgren, Sarah

2016-01-01

Feed-forward inhibitory (FFI) circuits are important for many information-processing functions. FFI circuit operations critically depend on the balance and timing between the excitatory and inhibitory components, which undergo rapid dynamic changes during neural activity due to short-term plasticity (STP) of both components. How dynamic changes in excitation/inhibition (E/I) balance during spike trains influence FFI circuit operations remains poorly understood. In the current study we examined the role of STP in the FFI circuit functions in the mouse hippocampus. Using a coincidence detection paradigm with simultaneous activation of two Schaffer collateral inputs, we found that the spiking probability in the target CA1 neuron was increased while spike precision concomitantly decreased during high-frequency bursts compared with a single spike. Blocking inhibitory synaptic transmission revealed that dynamics of inhibition predominately modulates the spike precision but not the changes in spiking probability, whereas the latter is modulated by the dynamics of excitation. Further analyses combining whole cell recordings and simulations of the FFI circuit suggested that dynamics of the inhibitory circuit component may influence spiking behavior during bursts by broadening the width of excitatory postsynaptic responses and that the strength of this modulation depends on the basal E/I ratio. We verified these predictions using a mouse model of fragile X syndrome, which has an elevated E/I ratio, and found a strongly reduced modulation of postsynaptic response width during bursts. Our results suggest that changes in the dynamics of excitatory and inhibitory circuit components due to STP play important yet distinct roles in modulating the properties of FFI circuits. PMID:27605532

Feedback and Feed-Forward for Promoting Problem-Based Learning in Online Learning Environments

ERIC Educational Resources Information Center

Webb, Ashley; Moallem, Mahnaz

2016-01-01

Purpose: The study aimed to (1) review the literature to construct conceptual models that could guide instructional designers in developing problem/project-based learning environments while applying effective feedback strategies, (2) use the models to design, develop, and implement an online graduate course, and (3) assess the efficiency of the…

Feed-forward and generalized regression neural networks in modeling feeding behavior of pigs in the grow-finishing phase

USDA-ARS?s Scientific Manuscript database

Feeding patterns in group-housed grow-finishing pigs have been investigated for use in management decisions, identifying sick animals, and determining genetic differences within a herd. Development of models to predict swine feeding behaviour has been limited due the large number of potential enviro...

Forecasting the portuguese stock market time series by using artificial neural networks

NASA Astrophysics Data System (ADS)

Isfan, Monica; Menezes, Rui; Mendes, Diana A.

2010-04-01

In this paper, we show that neural networks can be used to uncover the non-linearity that exists in the financial data. First, we follow a traditional approach by analysing the deterministic/stochastic characteristics of the Portuguese stock market data and some typical features are studied, like the Hurst exponents, among others. We also simulate a BDS test to investigate nonlinearities and the results are as expected: the financial time series do not exhibit linear dependence. Secondly, we trained four types of neural networks for the stock markets and used the models to make forecasts. The artificial neural networks were obtained using a three-layer feed-forward topology and the back-propagation learning algorithm. The quite large number of parameters that must be selected to develop a neural network forecasting model involves some trial and as a consequence the error is not small enough. In order to improve this we use a nonlinear optimization algorithm to minimize the error. Finally, the output of the 4 models is quite similar, leading to a qualitative forecast that we compare with the results of the application of k-nearest-neighbor for the same time series.

An ℋ∞ full information approach for the feedforward controller design of a large blended wing body flexible aircraft

NASA Astrophysics Data System (ADS)

Westermayer, C.; Schirrer, A.; Hemedi, M.; Kozek, M.

2013-12-01

An ℋ∞ full information feedforward design approach for longitudinal motion prefilter design of a large flexible blended wing body (BWB) aircraft is presented. An existing onset is extended such that specifications concerning command tracking, limited control energy, and manoeuvre load reduction can be addressed simultaneously. Therefore, the utilized design architecture is provided and manual tuning aspects are considered. In order to increase controller tuning efficiency, an automated tuning process based on several optimization criteria is proposed. Moreover, two design methodologies for the parameter-varying design case are investigated. The obtained controller is validated on a high-order nonlinear model, indicating the high potential of the presented approach for flexible aircraft control.

Quantum information processing with a travelling wave of light

NASA Astrophysics Data System (ADS)

Serikawa, Takahiro; Shiozawa, Yu; Ogawa, Hisashi; Takanashi, Naoto; Takeda, Shuntaro; Yoshikawa, Jun-ichi; Furusawa, Akira

2018-02-01

We exploit quantum information processing on a traveling wave of light, expecting emancipation from thermal noise, easy coupling to fiber communication, and potentially high operation speed. Although optical memories are technically challenging, we have an alternative approach to apply multi-step operations on traveling light, that is, continuous-variable one-way computation. So far our achievement includes generation of a one-million-mode entangled chain in time-domain, mode engineering of nonlinear resource states, and real-time nonlinear feedforward. Although they are implemented with free space optics, we are also investigating photonic integration and performed quantum teleportation with a passive liner waveguide chip as a demonstration of entangling, measurement, and feedforward. We also suggest a loop-based architecture as another model of continuous-variable computing.

Recent National Transonic Facility Test Process Improvements (Invited)

NASA Technical Reports Server (NTRS)

Kilgore, W. A.; Balakrishna, S.; Bobbitt, C. W., Jr.; Adcock, J. B.

2001-01-01

This paper describes the results of two recent process improvements; drag feed-forward Mach number control and simultaneous force/moment and pressure testing, at the National Transonic Facility. These improvements have reduced the duration and cost of testing. The drag feed-forward Mach number control reduces the Mach number settling time by using measured model drag in the Mach number control algorithm. Simultaneous force/moment and pressure testing allows simultaneous collection of force/moment and pressure data without sacrificing data quality thereby reducing the overall testing time. Both improvements can be implemented at any wind tunnel. Additionally the NTF is working to develop and implement continuous pitch as a testing option as an additional method to reduce costs and maintain data quality.

Recent National Transonic Facility Test Process Improvements (Invited)

NASA Technical Reports Server (NTRS)

Kilgore, W. A.; Balakrishna, S.; Bobbitt, C. W., Jr.; Adcock, J. B.

2001-01-01

This paper describes the results of two recent process improvements; drag feed-forward Mach number control and simultaneous force/moment and pressure testing, at the National Transonic Facility. These improvements have reduced the duration and cost of testing. The drag feedforward Mach number control reduces the Mach number settling time by using measured model drag in the Mach number control algorithm. Simultaneous force/moment and pressure testing allows simultaneous collection of force/moment and pressure data without sacrificing data quality thereby reducing the overall testing time. Both improvements can be implemented at any wind tunnel. Additionally the NTF is working to develop and implement continuous pitch as a testing option as an additional method to reduce costs and maintain data quality.

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

In silico prediction of drug-induced myelotoxicity by using Naïve Bayes method.

PubMed

Zhang, Hui; Yu, Peng; Zhang, Teng-Guo; Kang, Yan-Li; Zhao, Xiao; Li, Yuan-Yuan; He, Jia-Hui; Zhang, Ji

2015-11-01

Drug-induced myelotoxicity usually leads to decrease the production of platelets, red cells, and white cells. Thus, early identification and characterization of myelotoxicity hazard in drug development is very necessary. The purpose of this investigation was to develop a prediction model of drug-induced myelotoxicity by using a Naïve Bayes classifier. For comparison, other prediction models based on support vector machine and single-hidden-layer feed-forward neural network  methods were also established. Among all the prediction models, the Naïve Bayes classification model showed the best prediction performance, which offered an average overall prediction accuracy of [Formula: see text] for the training set and [Formula: see text] for the external test set. The significant contributions of this study are that we first developed a Naïve Bayes classification model of drug-induced myelotoxicity adverse effect using a larger scale dataset, which could be employed for the prediction of drug-induced myelotoxicity. In addition, several important molecular descriptors and substructures of myelotoxic compounds have been identified, which should be taken into consideration in the design of new candidate compounds to produce safer and more effective drugs, ultimately reducing the attrition rate in later stages of drug development.

Systematic comparison of the response properties of protein and RNA mediated gene regulatory motifs.

PubMed

Iyengar, Bharat Ravi; Pillai, Beena; Venkatesh, K V; Gadgil, Chetan J

2017-05-30

We present a framework enabling the dissection of the effects of motif structure (feedback or feedforward), the nature of the controller (RNA or protein), and the regulation mode (transcriptional, post-transcriptional or translational) on the response to a step change in the input. We have used a common model framework for gene expression where both motif structures have an activating input and repressing regulator, with the same set of parameters, to enable a comparison of the responses. We studied the global sensitivity of the system properties, such as steady-state gain, overshoot, peak time, and peak duration, to parameters. We find that, in all motifs, overshoot correlated negatively whereas peak duration varied concavely with peak time. Differences in the other system properties were found to be mainly dependent on the nature of the controller rather than the motif structure. Protein mediated motifs showed a higher degree of adaptation i.e. a tendency to return to baseline levels; in particular, feedforward motifs exhibited perfect adaptation. RNA mediated motifs had a mild regulatory effect; they also exhibited a lower peaking tendency and mean overshoot. Protein mediated feedforward motifs showed higher overshoot and lower peak time compared to the corresponding feedback motifs.

Automated optimal coordination of multiple-DOF neuromuscular actions in feedforward neuroprostheses.

PubMed

Lujan, J Luis; Crago, Patrick E

2009-01-01

This paper describes a new method for designing feedforward controllers for multiple-muscle, multiple-DOF, motor system neural prostheses. The design process is based on experimental measurement of the forward input/output properties of the neuromechanical system and numerical optimization of stimulation patterns to meet muscle coactivation criteria, thus resolving the muscle redundancy (i.e., overcontrol) and the coupled DOF problems inherent in neuromechanical systems. We designed feedforward controllers to control the isometric forces at the tip of the thumb in two directions during stimulation of three thumb muscles as a model system. We tested the method experimentally in ten able-bodied individuals and one patient with spinal cord injury. Good control of isometric force in both DOFs was observed, with rms errors less than 10% of the force range in seven experiments and statistically significant correlations between the actual and target forces in all ten experiments. Systematic bias and slope errors were observed in a few experiments, likely due to the neuromuscular fatigue. Overall, the tests demonstrated the ability of a general design approach to satisfy both control and coactivation criteria in multiple-muscle, multiple-axis neuromechanical systems, which is applicable to a wide range of neuromechanical systems and stimulation electrodes.

Density control in ITER: an iterative learning control and robust control approach

NASA Astrophysics Data System (ADS)

Ravensbergen, T.; de Vries, P. C.; Felici, F.; Blanken, T. C.; Nouailletas, R.; Zabeo, L.

2018-01-01

Plasma density control for next generation tokamaks, such as ITER, is challenging because of multiple reasons. The response of the usual gas valve actuators in future, larger fusion devices, might be too slow for feedback control. Both pellet fuelling and the use of feedforward-based control may help to solve this problem. Also, tight density limits arise during ramp-up, due to operational limits related to divertor detachment and radiative collapses. As the number of shots available for controller tuning will be limited in ITER, in this paper, iterative learning control (ILC) is proposed to determine optimal feedforward actuator inputs based on tracking errors, obtained in previous shots. This control method can take the actuator and density limits into account and can deal with large actuator delays. However, a purely feedforward-based density control may not be sufficient due to the presence of disturbances and shot-to-shot differences. Therefore, robust control synthesis is used to construct a robustly stabilizing feedback controller. In simulations, it is shown that this combined controller strategy is able to achieve good tracking performance in the presence of shot-to-shot differences, tight constraints, and model mismatches.

Feedback and feedforward locomotor adaptations to ankle-foot load in people with incomplete spinal cord injury.

PubMed

Gordon, Keith E; Wu, Ming; Kahn, Jennifer H; Schmit, Brian D

2010-09-01

Humans with spinal cord injury (SCI) modulate locomotor output in response to limb load. Understanding the neural control mechanisms responsible for locomotor adaptation could provide a framework for selecting effective interventions. We quantified feedback and feedforward locomotor adaptations to limb load modulations in people with incomplete SCI. While subjects airstepped (stepping performed with kinematic assistance and 100% bodyweight support), a powered-orthosis created a dorisflexor torque during the "stance phase" of select steps producing highly controlled ankle-load perturbations. When given repetitive, stance phase ankle-load, the increase in hip extension work, 0.27 J/kg above baseline (no ankle-load airstepping), was greater than the response to ankle-load applied during a single step, 0.14 J/kg (P = 0.029). This finding suggests that, at the hip, subjects produced both feedforward and feedback locomotor modulations. We estimate that, at the hip, the locomotor response to repetitive ankle-load was modulated almost equally by ongoing feedback and feedforward adaptations. The majority of subjects also showed after-effects in hip kinetic patterns that lasted 3 min in response to repetitive loading, providing additional evidence of feedforward locomotor adaptations. The magnitude of the after-effect was proportional to the response to repetitive ankle-foot load (R(2) = 0.92). In contrast, increases in soleus EMG amplitude were not different during repetitive and single-step ankle-load exposure, suggesting that ankle locomotor modulations were predominately feedback-based. Although subjects made both feedback and feedforward locomotor adaptations to changes in ankle-load, between-subject variations suggest that walking function may be related to the ability to make feedforward adaptations.

Digital controller design: Continuous and discrete describing function analysis of the IPS system

NASA Technical Reports Server (NTRS)

1977-01-01

The dynamic equations and the mathematical model of the continuous-data IPS control system are developed. The IPS model considered included one flexible body mode and was hardmounted to the Orbiter/Pallet. The model contains equations describing a torque feed-forward loop (using accelerometers as inputs) which will aid in reducing the pointing errors caused by Orbiter disturbances.

Pupil size directly modulates the feedforward response in human primary visual cortex independently of attention.

PubMed

Bombeke, Klaas; Duthoo, Wout; Mueller, Sven C; Hopf, Jens-Max; Boehler, C Nico

2016-02-15

Controversy revolves around the question of whether psychological factors like attention and emotion can influence the initial feedforward response in primary visual cortex (V1). Although traditionally, the electrophysiological correlate of this response in humans (the C1 component) has been found to be unaltered by psychological influences, a number of recent studies have described attentional and emotional modulations. Yet, research into psychological effects on the feedforward V1 response has neglected possible direct contributions of concomitant pupil-size modulations, which are known to also occur under various conditions of attentional load and emotional state. Here we tested the hypothesis that such pupil-size differences themselves directly affect the feedforward V1 response. We report data from two complementary experiments, in which we used procedures that modulate pupil size without differences in attentional load or emotion while simultaneously recording pupil-size and EEG data. Our results confirm that pupil size indeed directly influences the feedforward V1 response, showing an inverse relationship between pupil size and early V1 activity. While it is unclear in how far this effect represents a functionally-relevant adaptation, it identifies pupil-size differences as an important modulating factor of the feedforward response of V1 and could hence represent a confounding variable in research investigating the neural influence of psychological factors on early visual processing. Copyright © 2015 Elsevier Inc. All rights reserved.

A Feedforward Inhibitory Circuit Mediated by CB1-Expressing Fast-Spiking Interneurons in the Nucleus Accumbens.

PubMed

Wright, William J; Schlüter, Oliver M; Dong, Yan

2017-04-01

The nucleus accumbens (NAc) gates motivated behaviors through the functional output of principle medium spiny neurons (MSNs), whereas dysfunctional output of NAc MSNs contributes to a variety of psychiatric disorders. Fast-spiking interneurons (FSIs) are sparsely distributed throughout the NAc, forming local feedforward inhibitory circuits. It remains elusive how FSI-based feedforward circuits regulate the output of NAc MSNs. Here, we investigated a distinct subpopulation of NAc FSIs that express the cannabinoid receptor type-1 (CB1). Using a combination of paired electrophysiological recordings and pharmacological approaches, we characterized and compared feedforward inhibition of NAc MSNs from CB1 + FSIs and lateral inhibition from recurrent MSN collaterals. We observed that CB1 + FSIs exerted robust inhibitory control over a large percentage of nearby MSNs in contrast to local MSN collaterals that provided only sparse and weak inhibitory input to their neighboring MSNs. Furthermore, CB1 + FSI-mediated feedforward inhibition was preferentially suppressed by endocannabinoid (eCB) signaling, whereas MSN-mediated lateral inhibition was unaffected. Finally, we demonstrated that CB1 + FSI synapses onto MSNs are capable of undergoing experience-dependent long-term depression in a voltage- and eCB-dependent manner. These findings demonstrated that CB1 + FSIs are a major source of local inhibitory control of MSNs and a critical component of the feedforward inhibitory circuits regulating the output of the NAc.

A Feedforward Inhibitory Circuit Mediated by CB1-Expressing Fast-Spiking Interneurons in the Nucleus Accumbens

PubMed Central

Wright, William J; Schlüter, Oliver M; Dong, Yan

2017-01-01

The nucleus accumbens (NAc) gates motivated behaviors through the functional output of principle medium spiny neurons (MSNs), whereas dysfunctional output of NAc MSNs contributes to a variety of psychiatric disorders. Fast-spiking interneurons (FSIs) are sparsely distributed throughout the NAc, forming local feedforward inhibitory circuits. It remains elusive how FSI-based feedforward circuits regulate the output of NAc MSNs. Here, we investigated a distinct subpopulation of NAc FSIs that express the cannabinoid receptor type-1 (CB1). Using a combination of paired electrophysiological recordings and pharmacological approaches, we characterized and compared feedforward inhibition of NAc MSNs from CB1+ FSIs and lateral inhibition from recurrent MSN collaterals. We observed that CB1+ FSIs exerted robust inhibitory control over a large percentage of nearby MSNs in contrast to local MSN collaterals that provided only sparse and weak inhibitory input to their neighboring MSNs. Furthermore, CB1+ FSI-mediated feedforward inhibition was preferentially suppressed by endocannabinoid (eCB) signaling, whereas MSN-mediated lateral inhibition was unaffected. Finally, we demonstrated that CB1+ FSI synapses onto MSNs are capable of undergoing experience-dependent long-term depression in a voltage- and eCB-dependent manner. These findings demonstrated that CB1+ FSIs are a major source of local inhibitory control of MSNs and a critical component of the feedforward inhibitory circuits regulating the output of the NAc. PMID:27929113

Neural network pattern recognition of thermal-signature spectra for chemical defense

NASA Astrophysics Data System (ADS)

Carrieri, Arthur H.; Lim, Pascal I.

1995-05-01

We treat infrared patterns of absorption or emission by nerve and blister agent compounds (and simulants of this chemical group) as features for the training of neural networks to detect the compounds' liquid layers on the ground or their vapor plumes during evaporation by external heating. Training of a four-layer network architecture is composed of a backward-error-propagation algorithm and a gradient-descent paradigm. We conduct testing by feed-forwarding preprocessed spectra through the network in a scaled format consistent with the structure of the training-data-set representation. The best-performance weight matrix (spectral filter) evolved from final network training and testing with software simulation trials is electronically transferred to a set of eight artificial intelligence integrated circuits (ICs') in specific modular form (splitting of weight matrices). This form makes full use of all input-output IC nodes. This neural network computer serves an important real-time detection function when it is integrated into pre-and postprocessing data-handling units of a tactical prototype thermoluminescence sensor now under development at the Edgewood Research, Development, and Engineering Center.

Random Wiring, Ganglion Cell Mosaics, and the Functional Architecture of the Visual Cortex

PubMed Central

Coppola, David; White, Leonard E.; Wolf, Fred

2015-01-01

The architecture of iso-orientation domains in the primary visual cortex (V1) of placental carnivores and primates apparently follows species invariant quantitative laws. Dynamical optimization models assuming that neurons coordinate their stimulus preferences throughout cortical circuits linking millions of cells specifically predict these invariants. This might indicate that V1’s intrinsic connectome and its functional architecture adhere to a single optimization principle with high precision and robustness. To validate this hypothesis, it is critical to closely examine the quantitative predictions of alternative candidate theories. Random feedforward wiring within the retino-cortical pathway represents a conceptually appealing alternative to dynamical circuit optimization because random dimension-expanding projections are believed to generically exhibit computationally favorable properties for stimulus representations. Here, we ask whether the quantitative invariants of V1 architecture can be explained as a generic emergent property of random wiring. We generalize and examine the stochastic wiring model proposed by Ringach and coworkers, in which iso-orientation domains in the visual cortex arise through random feedforward connections between semi-regular mosaics of retinal ganglion cells (RGCs) and visual cortical neurons. We derive closed-form expressions for cortical receptive fields and domain layouts predicted by the model for perfectly hexagonal RGC mosaics. Including spatial disorder in the RGC positions considerably changes the domain layout properties as a function of disorder parameters such as position scatter and its correlations across the retina. However, independent of parameter choice, we find that the model predictions substantially deviate from the layout laws of iso-orientation domains observed experimentally. Considering random wiring with the currently most realistic model of RGC mosaic layouts, a pairwise interacting point process, the predicted layouts remain distinct from experimental observations and resemble Gaussian random fields. We conclude that V1 layout invariants are specific quantitative signatures of visual cortical optimization, which cannot be explained by generic random feedforward-wiring models. PMID:26575467

Neuromorphic Learning From Noisy Data

NASA Technical Reports Server (NTRS)

Merrill, Walter C.; Troudet, Terry

1993-01-01

Two reports present numerical study of performance of feedforward neural network trained by back-propagation algorithm in learning continuous-valued mappings from data corrupted by noise. Two types of noise considered: plant noise which affects dynamics of controlled process and data-processing noise, which occurs during analog processing and digital sampling of signals. Study performed with view toward use of neural networks as neurocontrollers to substitute for, or enhance, performances of human experts in controlling mechanical devices in presence of sensor and actuator noise and to enhance performances of more-conventional digital feedback electronic process controllers in noisy environments.

Causal feedforward control of a stochastically excited fuselage structure with active sidewall panel.

PubMed

Misol, Malte; Haase, Thomas; Monner, Hans Peter; Sinapius, Michael

2014-10-01

This paper provides experimental results of an aircraft-relevant double panel structure mounted in a sound transmission loss facility. The primary structure of the double panel system is excited either by a stochastic point force or by a diffuse sound field synthesized in the reverberation room of the transmission loss facility. The secondary structure, which is connected to the frames of the primary structure, is augmented by actuators and sensors implementing an active feedforward control system. Special emphasis is placed on the causality of the active feedforward control system and its implications on the disturbance rejection at the error sensors. The coherence of the sensor signals is analyzed for the two different disturbance excitations. Experimental results are presented regarding the causality, coherence, and disturbance rejection of the active feedforward control system. Furthermore, the sound transmission loss of the double panel system is evaluated for different configurations of the active system. A principal result of this work is the evidence that it is possible to strongly influence the transmission of stochastic disturbance sources through double panel configurations by means of an active feedforward control system.

Reinforcement learning of targeted movement in a spiking neuronal model of motor cortex.

PubMed

Chadderdon, George L; Neymotin, Samuel A; Kerr, Cliff C; Lytton, William W

2012-01-01

Sensorimotor control has traditionally been considered from a control theory perspective, without relation to neurobiology. In contrast, here we utilized a spiking-neuron model of motor cortex and trained it to perform a simple movement task, which consisted of rotating a single-joint "forearm" to a target. Learning was based on a reinforcement mechanism analogous to that of the dopamine system. This provided a global reward or punishment signal in response to decreasing or increasing distance from hand to target, respectively. Output was partially driven by Poisson motor babbling, creating stochastic movements that could then be shaped by learning. The virtual forearm consisted of a single segment rotated around an elbow joint, controlled by flexor and extensor muscles. The model consisted of 144 excitatory and 64 inhibitory event-based neurons, each with AMPA, NMDA, and GABA synapses. Proprioceptive cell input to this model encoded the 2 muscle lengths. Plasticity was only enabled in feedforward connections between input and output excitatory units, using spike-timing-dependent eligibility traces for synaptic credit or blame assignment. Learning resulted from a global 3-valued signal: reward (+1), no learning (0), or punishment (-1), corresponding to phasic increases, lack of change, or phasic decreases of dopaminergic cell firing, respectively. Successful learning only occurred when both reward and punishment were enabled. In this case, 5 target angles were learned successfully within 180 s of simulation time, with a median error of 8 degrees. Motor babbling allowed exploratory learning, but decreased the stability of the learned behavior, since the hand continued moving after reaching the target. Our model demonstrated that a global reinforcement signal, coupled with eligibility traces for synaptic plasticity, can train a spiking sensorimotor network to perform goal-directed motor behavior.

Sensorless control for permanent magnet synchronous motor using a neural network based adaptive estimator

NASA Astrophysics Data System (ADS)

Kwon, Chung-Jin; Kim, Sung-Joong; Han, Woo-Young; Min, Won-Kyoung

2005-12-01

The rotor position and speed estimation of permanent-magnet synchronous motor(PMSM) was dealt with. By measuring the phase voltages and currents of the PMSM drive, two diagonally recurrent neural network(DRNN) based observers, a neural current observer and a neural velocity observer were developed. DRNN which has self-feedback of the hidden neurons ensures that the outputs of DRNN contain the whole past information of the system even if the inputs of DRNN are only the present states and inputs of the system. Thus the structure of DRNN may be simpler than that of feedforward and fully recurrent neural networks. If the backpropagation method was used for the training of the DRNN the problem of slow convergence arise. In order to reduce this problem, recursive prediction error(RPE) based learning method for the DRNN was presented. The simulation results show that the proposed approach gives a good estimation of rotor speed and position, and RPE based training has requires a shorter computation time compared to backpropagation based training.

Electrotactile Feedback Improves Performance and Facilitates Learning in the Routine Grasping Task.

PubMed

Isaković, Milica; Belić, Minja; Štrbac, Matija; Popović, Igor; Došen, Strahinja; Farina, Dario; Keller, Thierry

2016-06-13

Aim of this study was to investigate the feasibility of electrotactile feedback in closed loop training of force control during the routine grasping task. The feedback was provided using an array electrode and a simple six-level spatial coding, and the experiment was conducted in three amputee subjects. The psychometric tests confirmed that the subjects could perceive and interpret the electrotactile feedback with a high success rate. The subjects performed the routine grasping task comprising 4 blocks of 60 grasping trials. In each trial, the subjects employed feedforward control to close the hand and produce the desired grasping force (four levels). First (baseline) and the last (validation) session were performed in open loop, while the second and the third session (training) included electrotactile feedback. The obtained results confirmed that using the feedback improved the accuracy and precision of the force control. In addition, the subjects performed significantly better in the validation vs. baseline session, therefore suggesting that electrotactile feedback can be used for learning and training of myoelectric control.

Automated corresponding point candidate selection for image registration using wavelet transformation neurla network with rotation invariant inputs and context information about neighboring candidates

NASA Astrophysics Data System (ADS)

Okumura, Hiroshi; Suezaki, Masashi; Sueyasu, Hideki; Arai, Kohei

2003-03-01

An automated method that can select corresponding point candidates is developed. This method has the following three features: 1) employment of the RIN-net for corresponding point candidate selection; 2) employment of multi resolution analysis with Haar wavelet transformation for improvement of selection accuracy and noise tolerance; 3) employment of context information about corresponding point candidates for screening of selected candidates. Here, the 'RIN-net' means the back-propagation trained feed-forward 3-layer artificial neural network that feeds rotation invariants as input data. In our system, pseudo Zernike moments are employed as the rotation invariants. The RIN-net has N x N pixels field of view (FOV). Some experiments are conducted to evaluate corresponding point candidate selection capability of the proposed method by using various kinds of remotely sensed images. The experimental results show the proposed method achieves fewer training patterns, less training time, and higher selection accuracy than conventional method.

A comparative study of breast cancer diagnosis based on neural network ensemble via improved training algorithms.

PubMed

Azami, Hamed; Escudero, Javier

2015-08-01

Breast cancer is one of the most common types of cancer in women all over the world. Early diagnosis of this kind of cancer can significantly increase the chances of long-term survival. Since diagnosis of breast cancer is a complex problem, neural network (NN) approaches have been used as a promising solution. Considering the low speed of the back-propagation (BP) algorithm to train a feed-forward NN, we consider a number of improved NN trainings for the Wisconsin breast cancer dataset: BP with momentum, BP with adaptive learning rate, BP with adaptive learning rate and momentum, Polak-Ribikre conjugate gradient algorithm (CGA), Fletcher-Reeves CGA, Powell-Beale CGA, scaled CGA, resilient BP (RBP), one-step secant and quasi-Newton methods. An NN ensemble, which is a learning paradigm to combine a number of NN outputs, is used to improve the accuracy of the classification task. Results demonstrate that NN ensemble-based classification methods have better performance than NN-based algorithms. The highest overall average accuracy is 97.68% obtained by NN ensemble trained by RBP for 50%-50% training-test evaluation method.

Turbidity forecasting at a karst spring using combined machine learning and wavelet multiresolution analysis.

NASA Astrophysics Data System (ADS)

Savary, M.; Massei, N.; Johannet, A.; Dupont, J. P.; Hauchard, E.

2016-12-01

25% of the world populations drink water extracted from karst aquifer. The comprehension and the protection of these aquifers appear as crucial due to an increase of drinking water needs. In Normandie(North-West of France), the principal exploited aquifer is the chalk aquifer. The chalk aquifer highly karstified is an important water resource, regionally speaking. Connections between surface and underground waters thanks to karstification imply turbidity that decreases water quality. Both numerous parameters and phenomenons, and the non-linearity of the rainfall/turbidity relation influence the turbidity causing difficulties to model and forecast turbidity peaks. In this context, the Yport pumping well provides half of Le Havreconurbation drinking water supply (236 000 inhabitants). The aim of this work is thus to perform prediction of the turbidity peaks in order to help pumping well managers to decrease the impact of turbidity on water treatment. Database consists in hourly rainfalls coming from six rain gauges located on the alimentation basin since 2009 and hourly turbidity since 1993. Because of the lack of accurate physical description of the karst system and its surface basin, the systemic paradigm is chosen and a black box model: a neural network model is chosen. In a first step, correlation analyses are used to design the original model architecture by identifying the relation between output and input. The following optimization phases bring us four different architectures. These models were experimented to forecast 12h ahead turbidity and threshold surpassing. The first model is a simple multilayer perceptron. The second is a two-branches model designed to better represent the fast (rainfall) and low (evapotranspiration) dynamics. Each kind of model is developed using both a recurrent and feed-forward architecture. This work highlights that feed-forward multilayer perceptron is better to predict turbidity peaks when feed-forward two-branches model is better to predict threshold surpassing. In a second step, the implementation of wavelet decomposition within the neural network model to better apprehend slow and fast dynamics is tested and discussed, which could also allows accounting for non-linearity of the turbid response to some extent. This second approach is still under realization so far.

Automated electronic tongue based on potentiometric sensors for the determination of a trinary anionic surfactant mixture.

PubMed

Cortina, Montserrat; Ecker, Christina; Calvo, Daniel; del Valle, Manuel

2008-01-22

An automated electronic tongue consisting of an array of potentiometric sensors and an artificial neural network (ANN) has been developed to resolve mixtures of anionic surfactants. The sensor array was formed by five different flow-through sensors for anionic surfactants, based on poly(vinyl chloride) membranes having cross-sensitivity features. Feedforward multilayer neural networks were used to predict surfactant concentrations. As a great amount of information is required for the correct modelling of the sensors response, a sequential injection analysis (SIA) system was used to automatically provide it. Dodecylsulfate (DS(-)), dodecylbenzenesulfonate (DBS(-)) and alpha-alkene sulfonate (ALF(-)) formed the three-analyte study case resolved in this work. Their concentrations varied from 0.2 to 4mM for ALF(-) and DBS(-) and from 0.2 to 5mM for DS(-). Good prediction ability was obtained with correlation coefficients better than 0.933 when the obtained values were compared with those expected for a set of 16 external test samples not used for training.

Perceptual learning as improved probabilistic inference in early sensory areas.

PubMed

Bejjanki, Vikranth R; Beck, Jeffrey M; Lu, Zhong-Lin; Pouget, Alexandre

2011-05-01

Extensive training on simple tasks such as fine orientation discrimination results in large improvements in performance, a form of learning known as perceptual learning. Previous models have argued that perceptual learning is due to either sharpening and amplification of tuning curves in early visual areas or to improved probabilistic inference in later visual areas (at the decision stage). However, early theories are inconsistent with the conclusions of psychophysical experiments manipulating external noise, whereas late theories cannot explain the changes in neural responses that have been reported in cortical areas V1 and V4. Here we show that we can capture both the neurophysiological and behavioral aspects of perceptual learning by altering only the feedforward connectivity in a recurrent network of spiking neurons so as to improve probabilistic inference in early visual areas. The resulting network shows modest changes in tuning curves, in line with neurophysiological reports, along with a marked reduction in the amplitude of pairwise noise correlations.

Landsat 7 thermal-IR image sharpening using an artificial neural network and sensor model

USGS Publications Warehouse

Lemeshewsky, G.P.; Schowengerdt, R.A.; ,

2001-01-01

The enhanced thematic mapper (plus) (ETM+) instrument on Landsat 7 shares the same basic design as the TM sensors on Landsats 4 and 5, with some significant improvements. In common are six multispectral bands with a 30-m ground-projected instantaneous field of view (GIFOV). However, the thermaL-IR (TIR) band now has a 60-m GIFOV, instead of 120-m. Also, a 15-m panchromatic band has been added. The artificial neural network (NN) image sharpening method described here uses data from the higher spatial resolution ETM+ bands to enhance (sharpen) the spatial resolution of the TIR imagery. It is based on an assumed correlation over multiple scales of resolution, between image edge contrast patterns in the TIR band and several other spectral bands. A multilayer, feedforward NN is trained to approximate TIR data at 60m, given degraded (from 30-m to 60-m) spatial resolution input from spectral bands 7, 5, and 2. After training, the NN output for full-resolution input generates an approximation of a TIR image at 30-m resolution. Two methods are used to degrade the spatial resolution of the imagery used for NN training, and the corresponding sharpening results are compared. One degradation method uses a published sensor transfer function (TF) for Landsat 5 to simulate sensor coarser resolution imagery from higher resolution imagery. For comparison, the second degradation method is simply Gaussian low pass filtering and subsampling, wherein the Gaussian filter approximates the full width at half maximum amplitude characteristics of the TF-based spatial filter. Two fixed-size NNs (that is, number of weights and processing elements) were trained separately with the degraded resolution data, and the sharpening results compared. The comparison evaluates the relative influence of the degradation technique employed and whether or not it is desirable to incorporate a sensor TF model. Preliminary results indicate some improvements for the sensor model-based technique. Further evaluation using a higher resolution reference image and strict application of sensor model to data is recommended.

A new rate-dependent model for high-frequency tracking performance enhancement of piezoactuator system

NASA Astrophysics Data System (ADS)

Tian, Lizhi; Xiong, Zhenhua; Wu, Jianhua; Ding, Han

2017-05-01

Feedforward-feedback control is widely used in motion control of piezoactuator systems. Due to the phase lag caused by incomplete dynamics compensation, the performance of the composite controller is greatly limited at high frequency. This paper proposes a new rate-dependent model to improve the high-frequency tracking performance by reducing dynamics compensation error. The rate-dependent model is designed as a function of the input and input variation rate to describe the input-output relationship of the residual system dynamics which mainly performs as phase lag in a wide frequency band. Then the direct inversion of the proposed rate-dependent model is used to compensate the residual system dynamics. Using the proposed rate-dependent model as feedforward term, the open loop performance can be improved significantly at medium-high frequency. Then, combining the with feedback controller, the composite controller can provide enhanced close loop performance from low frequency to high frequency. At the frequency of 1 Hz, the proposed controller presents the same performance as previous methods. However, at the frequency of 900 Hz, the tracking error is reduced to be 30.7% of the decoupled approach.

Direct process estimation from tomographic data using artificial neural systems

NASA Astrophysics Data System (ADS)

Mohamad-Saleh, Junita; Hoyle, Brian S.; Podd, Frank J.; Spink, D. M.

2001-07-01

The paper deals with the goal of component fraction estimation in multicomponent flows, a critical measurement in many processes. Electrical capacitance tomography (ECT) is a well-researched sensing technique for this task, due to its low-cost, non-intrusion, and fast response. However, typical systems, which include practicable real-time reconstruction algorithms, give inaccurate results, and existing approaches to direct component fraction measurement are flow-regime dependent. In the investigation described, an artificial neural network approach is used to directly estimate the component fractions in gas-oil, gas-water, and gas-oil-water flows from ECT measurements. A 2D finite- element electric field model of a 12-electrode ECT sensor is used to simulate ECT measurements of various flow conditions. The raw measurements are reduced to a mutually independent set using principal components analysis and used with their corresponding component fractions to train multilayer feed-forward neural networks (MLFFNNs). The trained MLFFNNs are tested with patterns consisting of unlearned ECT simulated and plant measurements. Results included in the paper have a mean absolute error of less than 1% for the estimation of various multicomponent fractions of the permittivity distribution. They are also shown to give improved component fraction estimation compared to a well known direct ECT method.

New recursive-least-squares algorithms for nonlinear active control of sound and vibration using neural networks.

PubMed

Bouchard, M

2001-01-01

In recent years, a few articles describing the use of neural networks for nonlinear active control of sound and vibration were published. Using a control structure with two multilayer feedforward neural networks (one as a nonlinear controller and one as a nonlinear plant model), steepest descent algorithms based on two distinct gradient approaches were introduced for the training of the controller network. The two gradient approaches were sometimes called the filtered-x approach and the adjoint approach. Some recursive-least-squares algorithms were also introduced, using the adjoint approach. In this paper, an heuristic procedure is introduced for the development of recursive-least-squares algorithms based on the filtered-x and the adjoint gradient approaches. This leads to the development of new recursive-least-squares algorithms for the training of the controller neural network in the two networks structure. These new algorithms produce a better convergence performance than previously published algorithms. Differences in the performance of algorithms using the filtered-x and the adjoint gradient approaches are discussed in the paper. The computational load of the algorithms discussed in the paper is evaluated for multichannel systems of nonlinear active control. Simulation results are presented to compare the convergence performance of the algorithms, showing the convergence gain provided by the new algorithms.

Correcting wave predictions with artificial neural networks

NASA Astrophysics Data System (ADS)

Makarynskyy, O.; Makarynska, D.

2003-04-01

The predictions of wind waves with different lead times are necessary in a large scope of coastal and open ocean activities. Numerical wave models, which usually provide this information, are based on deterministic equations that do not entirely account for the complexity and uncertainty of the wave generation and dissipation processes. An attempt to improve wave parameters short-term forecasts based on artificial neural networks is reported. In recent years, artificial neural networks have been used in a number of coastal engineering applications due to their ability to approximate the nonlinear mathematical behavior without a priori knowledge of interrelations among the elements within a system. The common multilayer feed-forward networks, with a nonlinear transfer functions in the hidden layers, were developed and employed to forecast the wave characteristics over one hour intervals starting from one up to 24 hours, and to correct these predictions. Three non-overlapping data sets of wave characteristics, both from a buoy, moored roughly 60 miles west of the Aran Islands, west coast of Ireland, were used to train and validate the neural nets involved. The networks were trained with error back propagation algorithm. Time series plots and scatterplots of the wave characteristics as well as tables with statistics show an improvement of the results achieved due to the correction procedure employed.

A comparative analysis of support vector machines and extreme learning machines.

PubMed

Liu, Xueyi; Gao, Chuanhou; Li, Ping

2012-09-01

The theory of extreme learning machines (ELMs) has recently become increasingly popular. As a new learning algorithm for single-hidden-layer feed-forward neural networks, an ELM offers the advantages of low computational cost, good generalization ability, and ease of implementation. Hence the comparison and model selection between ELMs and other kinds of state-of-the-art machine learning approaches has become significant and has attracted many research efforts. This paper performs a comparative analysis of the basic ELMs and support vector machines (SVMs) from two viewpoints that are different from previous works: one is the Vapnik-Chervonenkis (VC) dimension, and the other is their performance under different training sample sizes. It is shown that the VC dimension of an ELM is equal to the number of hidden nodes of the ELM with probability one. Additionally, their generalization ability and computational complexity are exhibited with changing training sample size. ELMs have weaker generalization ability than SVMs for small sample but can generalize as well as SVMs for large sample. Remarkably, great superiority in computational speed especially for large-scale sample problems is found in ELMs. The results obtained can provide insight into the essential relationship between them, and can also serve as complementary knowledge for their past experimental and theoretical comparisons. Copyright © 2012 Elsevier Ltd. All rights reserved.

A feed-forward neural network for modeling feeding behavior of group-housed grow-finish pigs with respect to thermal conditions

USDA-ARS?s Scientific Manuscript database

Feeding patterns of pigs in the grow-finish phase have been investigated for use in management decisions, identifying sick animals, and determining genetic differences within a herd. Development of models to predict swine feeding behavior has been limited due the large number of potential environmen...

Minimal-Inversion Feedforward-And-Feedback Control System

NASA Technical Reports Server (NTRS)

Seraji, Homayoun

1990-01-01

Recent developments in theory of control systems support concept of minimal-inversion feedforward-and feedback control system consisting of three independently designable control subsystems. Applicable to the control of linear, time-invariant plant.

Feedforward somatosensory inhibition is normal in cervical dystonia.

PubMed

Ferrè, Elisa R; Ganos, Christos; Bhatia, Kailash P; Haggard, Patrick

2015-03-01

Insufficient cortical inhibition is a key pathophysiological finding in dystonia. Subliminal sensory stimuli were reported to transiently inhibit somatosensory processing. Here we investigated whether such subliminal feedforward inhibition is reduced in patients with cervical dystonia. Sixteen cervical dystonia patients and 16 matched healthy controls performed a somatosensory detection task. We measured the drop in sensitivity to detect a threshold-level digital nerve shock when it was preceded by a subliminal conditioning shock, compared to when it was not. Subliminal conditioning shocks reduced sensitivity to threshold stimuli to a similar extent in both patients and controls, suggesting that somatosensory subliminal feedforward inhibition is normal in cervical dystonia. Somatosensory feedforward inhibition was normal in this group of cervical dystonia patients. Our results qualify previous concepts of a general dystonic deficit in sensorimotor inhibitory processing. Copyright © 2015 Elsevier Ltd. All rights reserved.

Structural learning in feedforward and feedback control.

PubMed

Yousif, Nada; Diedrichsen, Jörn

2012-11-01

For smooth and efficient motor control, the brain needs to make fast corrections during the movement to resist possible perturbations. It also needs to adapt subsequent movements to improve future performance. It is important that both feedback corrections and feedforward adaptation need to be made based on noisy and often ambiguous sensory data. Therefore, the initial response of the motor system, both for online corrections and adaptive responses, is guided by prior assumptions about the likely structure of perturbations. In the context of correcting and adapting movements perturbed by a force field, we asked whether these priors are hard wired or whether they can be modified through repeated exposure to differently shaped force fields. We found that both feedback corrections to unexpected perturbations and feedforward adaptation to a new force field changed, such that they were appropriate to counteract the type of force field that participants had experienced previously. We then investigated whether these changes were driven by a common mechanism or by two separate mechanisms. Participants experienced force fields that were either temporally consistent, causing sustained adaptation, or temporally inconsistent, causing little overall adaptation. We found that the consistent force fields modified both feedback and feedforward responses. In contrast, the inconsistent force field modified the temporal shape of feedback corrections but not of the feedforward adaptive response. These results indicate that responses to force perturbations can be modified in a structural manner and that these modifications are at least partly dissociable for feedback and feedforward control.

Structural learning in feedforward and feedback control

PubMed Central

Diedrichsen, Jörn

2012-01-01

For smooth and efficient motor control, the brain needs to make fast corrections during the movement to resist possible perturbations. It also needs to adapt subsequent movements to improve future performance. It is important that both feedback corrections and feedforward adaptation need to be made based on noisy and often ambiguous sensory data. Therefore, the initial response of the motor system, both for online corrections and adaptive responses, is guided by prior assumptions about the likely structure of perturbations. In the context of correcting and adapting movements perturbed by a force field, we asked whether these priors are hard wired or whether they can be modified through repeated exposure to differently shaped force fields. We found that both feedback corrections to unexpected perturbations and feedforward adaptation to a new force field changed, such that they were appropriate to counteract the type of force field that participants had experienced previously. We then investigated whether these changes were driven by a common mechanism or by two separate mechanisms. Participants experienced force fields that were either temporally consistent, causing sustained adaptation, or temporally inconsistent, causing little overall adaptation. We found that the consistent force fields modified both feedback and feedforward responses. In contrast, the inconsistent force field modified the temporal shape of feedback corrections but not of the feedforward adaptive response. These results indicate that responses to force perturbations can be modified in a structural manner and that these modifications are at least partly dissociable for feedback and feedforward control. PMID:22896725

Autoshaped choice in artificial neural networks: implications for behavioral economics and neuroeconomics.

PubMed

Burgos, José E; García-Leal, Óscar

2015-05-01

An existing neural network model of conditioning was used to simulate autoshaped choice. In this phenomenon, pigeons first receive an autoshaping procedure with two keylight stimuli X and Y separately paired with food in a forward-delay manner, intermittently for X and continuously for Y. Then pigeons receive unreinforced choice test trials of X and Y concurrently present. Most pigeons choose Y. This preference for a more valuable response alternative is a form of economic behavior that makes the phenomenon relevant to behavioral economics. The phenomenon also suggests a role for Pavlovian contingencies in economic behavior. The model used, in contrast to others, predicts autoshaping and automaintenance, so it is uniquely positioned to predict autoshaped choice. The model also contemplates neural substrates of economic behavior in neuroeconomics, such as dopaminergic and hippocampal systems. A feedforward neural network architecture was designed to simulate a neuroanatomical differentiation between two environment-behavior relations X-R1 and Y-R2, [corrected] where R1 and R2 denote two different emitted responses (not unconditionally elicited by the reward). Networks with this architecture received a training protocol that simulated an autoshaped-choice procedure. Most networks simulated the phenomenon. Implications for behavioral economics and neuroeconomics, limitations, and the issue of model appraisal are discussed. Copyright © 2015 Elsevier B.V. All rights reserved.

Tool Steel Heat Treatment Optimization Using Neural Network Modeling

NASA Astrophysics Data System (ADS)

Podgornik, Bojan; Belič, Igor; Leskovšek, Vojteh; Godec, Matjaz

2016-11-01

Optimization of tool steel properties and corresponding heat treatment is mainly based on trial and error approach, which requires tremendous experimental work and resources. Therefore, there is a huge need for tools allowing prediction of mechanical properties of tool steels as a function of composition and heat treatment process variables. The aim of the present work was to explore the potential and possibilities of artificial neural network-based modeling to select and optimize vacuum heat treatment conditions depending on the hot work tool steel composition and required properties. In the current case training of the feedforward neural network with error backpropagation training scheme and four layers of neurons (8-20-20-2) scheme was based on the experimentally obtained tempering diagrams for ten different hot work tool steel compositions and at least two austenitizing temperatures. Results show that this type of modeling can be successfully used for detailed and multifunctional analysis of different influential parameters as well as to optimize heat treatment process of hot work tool steels depending on the composition. In terms of composition, V was found as the most beneficial alloying element increasing hardness and fracture toughness of hot work tool steel; Si, Mn, and Cr increase hardness but lead to reduced fracture toughness, while Mo has the opposite effect. Optimum concentration providing high KIc/HRC ratios would include 0.75 pct Si, 0.4 pct Mn, 5.1 pct Cr, 1.5 pct Mo, and 0.5 pct V, with the optimum heat treatment performed at lower austenitizing and intermediate tempering temperatures.

The role of multisensor data fusion in neuromuscular control of a sagittal arm with a pair of muscles using actor-critic reinforcement learning method.

PubMed

Golkhou, V; Parnianpour, M; Lucas, C

2004-01-01

In this study, we consider the role of multisensor data fusion in neuromuscular control using an actor-critic reinforcement learning method. The model we use is a single link system actuated by a pair of muscles that are excited with alpha and gamma signals. Various physiological sensor information such as proprioception, spindle sensors, and Golgi tendon organs have been integrated to achieve an oscillatory movement with variable amplitude and frequency, while achieving a stable movement with minimum metabolic cost and coactivation. The system is highly nonlinear in all its physical and physiological attributes. Transmission delays are included in the afferent and efferent neural paths to account for a more accurate representation of the reflex loops. This paper proposes a reinforcement learning method with an Actor-Critic architecture instead of middle and low level of central nervous system (CNS). The Actor in this structure is a two layer feedforward neural network and the Critic is a model of the cerebellum. The Critic is trained by the State-Action-Reward-State-Action (SARSA) method. The Critic will train the Actor by supervisory learning based on previous experiences. The reinforcement signal in SARSA is evaluated based on available alternatives concerning the concept of multisensor data fusion. The effectiveness and the biological plausibility of the present model are demonstrated by several simulations. The system showed excellent tracking capability when we integrated the available sensor information. Addition of a penalty for activation of muscles resulted in much lower muscle coactivation while keeping the movement stable.

Feedback and feedforward adaptation to visuomotor delay during reaching and slicing movements.

PubMed

Botzer, Lior; Karniel, Amir

2013-07-01

It has been suggested that the brain and in particular the cerebellum and motor cortex adapt to represent the environment during reaching movements under various visuomotor perturbations. It is well known that significant delay is present in neural conductance and processing; however, the possible representation of delay and adaptation to delayed visual feedback has been largely overlooked. Here we investigated the control of reaching movements in human subjects during an imposed visuomotor delay in a virtual reality environment. In the first experiment, when visual feedback was unexpectedly delayed, the hand movement overshot the end-point target, indicating a vision-based feedback control. Over the ensuing trials, movements gradually adapted and became accurate. When the delay was removed unexpectedly, movements systematically undershot the target, demonstrating that adaptation occurred within the vision-based feedback control mechanism. In a second experiment designed to broaden our understanding of the underlying mechanisms, we revealed similar after-effects for rhythmic reversal (out-and-back) movements. We present a computational model accounting for these results based on two adapted forward models, each tuned for a specific modality delay (proprioception or vision), and a third feedforward controller. The computational model, along with the experimental results, refutes delay representation in a pure forward vision-based predictor and suggests that adaptation occurred in the forward vision-based predictor, and concurrently in the state-based feedforward controller. Understanding how the brain compensates for conductance and processing delays is essential for understanding certain impairments concerning these neural delays as well as for the development of brain-machine interfaces. © 2013 Federation of European Neuroscience Societies and John Wiley & Sons Ltd.

Short-term Synaptic Depression in the Feedforward Inhibitory Circuit in the Dorsal Lateral Geniculate Nucleus.

PubMed

Augustinaite, Sigita; Heggelund, Paul

2018-05-24

Synaptic short-term plasticity (STP) regulates synaptic transmission in an activity-dependent manner and thereby has important roles in the signal processing in the brain. In some synapses, a presynaptic train of action potentials elicits post-synaptic potentials that gradually increase during the train (facilitation), but in other synapses, these potentials gradually decrease (depression). We studied STP in neurons in the visual thalamic relay, the dorsal lateral geniculate nucleus (dLGN). The dLGN contains two types of neurons: excitatory thalamocortical (TC) neurons, which transfer signals from retinal afferents to visual cortex, and local inhibitory interneurons, which form an inhibitory feedforward loop that regulates the thalamocortical signal transmission. The overall STP in the retino-thalamic relay is short-term depression, but the distinct kind and characteristics of the plasticity at the different types of synapses are unknown. We studied STP in the excitatory responses of interneurons to stimulation of retinal afferents, in the inhibitory responses of TC neurons to stimulation of afferents from interneurons, and in the disynaptic inhibitory responses of TC neurons to stimulation of retinal afferents. Moreover, we studied STP at the direct excitatory input to TC neurons from retinal afferents. The STP at all types of the synapses showed short-term depression. This depression can accentuate rapid changes in the stream of signals and thereby promote detectability of significant features in the sensory input. In vision, detection of edges and contours is essential for object perception, and the synaptic short-term depression in the early visual pathway provides important contributions to this detection process. Copyright © 2018 IBRO. Published by Elsevier Ltd. All rights reserved.

The predictability of frequency-altered auditory feedback changes the weighting of feedback and feedforward input for speech motor control.

PubMed

Scheerer, Nichole E; Jones, Jeffery A

2014-12-01

Speech production requires the combined effort of a feedback control system driven by sensory feedback, and a feedforward control system driven by internal models. However, the factors that dictate the relative weighting of these feedback and feedforward control systems are unclear. In this event-related potential (ERP) study, participants produced vocalisations while being exposed to blocks of frequency-altered feedback (FAF) perturbations that were either predictable in magnitude (consistently either 50 or 100 cents) or unpredictable in magnitude (50- and 100-cent perturbations varying randomly within each vocalisation). Vocal and P1-N1-P2 ERP responses revealed decreases in the magnitude and trial-to-trial variability of vocal responses, smaller N1 amplitudes, and shorter vocal, P1 and N1 response latencies following predictable FAF perturbation magnitudes. In addition, vocal response magnitudes correlated with N1 amplitudes, vocal response latencies, and P2 latencies. This pattern of results suggests that after repeated exposure to predictable FAF perturbations, the contribution of the feedforward control system increases. Examination of the presentation order of the FAF perturbations revealed smaller compensatory responses, smaller P1 and P2 amplitudes, and shorter N1 latencies when the block of predictable 100-cent perturbations occurred prior to the block of predictable 50-cent perturbations. These results suggest that exposure to large perturbations modulates responses to subsequent perturbations of equal or smaller size. Similarly, exposure to a 100-cent perturbation prior to a 50-cent perturbation within a vocalisation decreased the magnitude of vocal and N1 responses, but increased P1 and P2 latencies. Thus, exposure to a single perturbation can affect responses to subsequent perturbations. © 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd.

Toward heterogeneity in feedforward network with synaptic delays based on FitzHugh-Nagumo model

NASA Astrophysics Data System (ADS)

Qin, Ying-Mei; Men, Cong; Zhao, Jia; Han, Chun-Xiao; Che, Yan-Qiu

2018-01-01

We focus on the role of heterogeneity on the propagation of firing patterns in feedforward network (FFN). Effects of heterogeneities both in parameters of neuronal excitability and synaptic delays are investigated systematically. Neuronal heterogeneity is found to modulate firing rates and spiking regularity by changing the excitability of the network. Synaptic delays are strongly related with desynchronized and synchronized firing patterns of the FFN, which indicate that synaptic delays may play a significant role in bridging rate coding and temporal coding. Furthermore, quasi-coherence resonance (quasi-CR) phenomenon is observed in the parameter domain of connection probability and delay-heterogeneity. All these phenomena above enable a detailed characterization of neuronal heterogeneity in FFN, which may play an indispensable role in reproducing the important properties of in vivo experiments.

Propagation of spiking regularity and double coherence resonance in feedforward networks.

PubMed

Men, Cong; Wang, Jiang; Qin, Ying-Mei; Deng, Bin; Tsang, Kai-Ming; Chan, Wai-Lok

2012-03-01

We investigate the propagation of spiking regularity in noisy feedforward networks (FFNs) based on FitzHugh-Nagumo neuron model systematically. It is found that noise could modulate the transmission of firing rate and spiking regularity. Noise-induced synchronization and synfire-enhanced coherence resonance are also observed when signals propagate in noisy multilayer networks. It is interesting that double coherence resonance (DCR) with the combination of synaptic input correlation and noise intensity is finally attained after the processing layer by layer in FFNs. Furthermore, inhibitory connections also play essential roles in shaping DCR phenomena. Several properties of the neuronal network such as noise intensity, correlation of synaptic inputs, and inhibitory connections can serve as control parameters in modulating both rate coding and the order of temporal coding.

Clinical application of modified bag-of-features coupled with hybrid neural-based classifier in dengue fever classification using gene expression data.

PubMed

Chatterjee, Sankhadeep; Dey, Nilanjan; Shi, Fuqian; Ashour, Amira S; Fong, Simon James; Sen, Soumya

2018-04-01

Dengue fever detection and classification have a vital role due to the recent outbreaks of different kinds of dengue fever. Recently, the advancement in the microarray technology can be employed for such classification process. Several studies have established that the gene selection phase takes a significant role in the classifier performance. Subsequently, the current study focused on detecting two different variations, namely, dengue fever (DF) and dengue hemorrhagic fever (DHF). A modified bag-of-features method has been proposed to select the most promising genes in the classification process. Afterward, a modified cuckoo search optimization algorithm has been engaged to support the artificial neural (ANN-MCS) to classify the unknown subjects into three different classes namely, DF, DHF, and another class containing convalescent and normal cases. The proposed method has been compared with other three well-known classifiers, namely, multilayer perceptron feed-forward network (MLP-FFN), artificial neural network (ANN) trained with cuckoo search (ANN-CS), and ANN trained with PSO (ANN-PSO). Experiments have been carried out with different number of clusters for the initial bag-of-features-based feature selection phase. After obtaining the reduced dataset, the hybrid ANN-MCS model has been employed for the classification process. The results have been compared in terms of the confusion matrix-based performance measuring metrics. The experimental results indicated a highly statistically significant improvement with the proposed classifier over the traditional ANN-CS model.

Effective Data-Driven Calibration for a Galvanometric Laser Scanning System Using Binocular Stereo Vision.

PubMed

Tu, Junchao; Zhang, Liyan

2018-01-12

A new solution to the problem of galvanometric laser scanning (GLS) system calibration is presented. Under the machine learning framework, we build a single-hidden layer feedforward neural network (SLFN）to represent the GLS system, which takes the digital control signal at the drives of the GLS system as input and the space vector of the corresponding outgoing laser beam as output. The training data set is obtained with the aid of a moving mechanism and a binocular stereo system. The parameters of the SLFN are efficiently solved in a closed form by using extreme learning machine (ELM). By quantitatively analyzing the regression precision with respective to the number of hidden neurons in the SLFN, we demonstrate that the proper number of hidden neurons can be safely chosen from a broad interval to guarantee good generalization performance. Compared to the traditional model-driven calibration, the proposed calibration method does not need a complex modeling process and is more accurate and stable. As the output of the network is the space vectors of the outgoing laser beams, it costs much less training time and can provide a uniform solution to both laser projection and 3D-reconstruction, in contrast with the existing data-driven calibration method which only works for the laser triangulation problem. Calibration experiment, projection experiment and 3D reconstruction experiment are respectively conducted to test the proposed method, and good results are obtained.

Feature to prototype transition in neural networks

NASA Astrophysics Data System (ADS)

Krotov, Dmitry; Hopfield, John

Models of associative memory with higher order (higher than quadratic) interactions, and their relationship to neural networks used in deep learning are discussed. Associative memory is conventionally described by recurrent neural networks with dynamical convergence to stable points. Deep learning typically uses feedforward neural nets without dynamics. However, a simple duality relates these two different views when applied to problems of pattern classification. From the perspective of associative memory such models deserve attention because they make it possible to store a much larger number of memories, compared to the quadratic case. In the dual description, these models correspond to feedforward neural networks with one hidden layer and unusual activation functions transmitting the activities of the visible neurons to the hidden layer. These activation functions are rectified polynomials of a higher degree rather than the rectified linear functions used in deep learning. The network learns representations of the data in terms of features for rectified linear functions, but as the power in the activation function is increased there is a gradual shift to a prototype-based representation, the two extreme regimes of pattern recognition known in cognitive psychology. Simons Center for Systems Biology.

Neural Sequence Generation Using Spatiotemporal Patterns of Inhibition.

PubMed

Cannon, Jonathan; Kopell, Nancy; Gardner, Timothy; Markowitz, Jeffrey

2015-11-01

Stereotyped sequences of neural activity are thought to underlie reproducible behaviors and cognitive processes ranging from memory recall to arm movement. One of the most prominent theoretical models of neural sequence generation is the synfire chain, in which pulses of synchronized spiking activity propagate robustly along a chain of cells connected by highly redundant feedforward excitation. But recent experimental observations in the avian song production pathway during song generation have shown excitatory activity interacting strongly with the firing patterns of inhibitory neurons, suggesting a process of sequence generation more complex than feedforward excitation. Here we propose a model of sequence generation inspired by these observations in which a pulse travels along a spatially recurrent excitatory chain, passing repeatedly through zones of local feedback inhibition. In this model, synchrony and robust timing are maintained not through redundant excitatory connections, but rather through the interaction between the pulse and the spatiotemporal pattern of inhibition that it creates as it circulates the network. These results suggest that spatially and temporally structured inhibition may play a key role in sequence generation.

Finite element modelling of human auditory periphery including a feed-forward amplification of the cochlea.

PubMed

Wang, Xuelin; Wang, Liling; Zhou, Jianjun; Hu, Yujin

2014-08-01

A three-dimensional finite element model is developed for the simulation of the sound transmission through the human auditory periphery consisting of the external ear canal, middle ear and cochlea. The cochlea is modelled as a straight duct divided into two fluid-filled scalae by the basilar membrane (BM) having an orthotropic material property with dimensional variation along its length. In particular, an active feed-forward mechanism is added into the passive cochlear model to represent the activity of the outer hair cells (OHCs). An iterative procedure is proposed for calculating the nonlinear response resulting from the active cochlea in the frequency domain. Results on the middle-ear transfer function, BM steady-state frequency response and intracochlear pressure are derived. A good match of the model predictions with experimental data from the literatures demonstrates the validity of the ear model for simulating sound pressure gain of middle ear, frequency to place map, cochlear sensitivity and compressive output for large intensity input. The current model featuring an active cochlea is able to correlate directly the sound stimulus in the ear canal with the vibration of BM and provides a tool to explore the mechanisms by which sound pressure in the ear canal is converted to a stimulus for the OHCs.

The inhibitory microcircuit of the substantia nigra provides feedback gain control of the basal ganglia output

PubMed Central

Brown, Jennifer; Pan, Wei-Xing; Dudman, Joshua Tate

2014-01-01

Dysfunction of the basal ganglia produces severe deficits in the timing, initiation, and vigor of movement. These diverse impairments suggest a control system gone awry. In engineered systems, feedback is critical for control. By contrast, models of the basal ganglia highlight feedforward circuitry and ignore intrinsic feedback circuits. In this study, we show that feedback via axon collaterals of substantia nigra projection neurons control the gain of the basal ganglia output. Through a combination of physiology, optogenetics, anatomy, and circuit mapping, we elaborate a general circuit mechanism for gain control in a microcircuit lacking interneurons. Our data suggest that diverse tonic firing rates, weak unitary connections and a spatially diffuse collateral circuit with distinct topography and kinetics from feedforward input is sufficient to implement divisive feedback inhibition. The importance of feedback for engineered systems implies that the intranigral microcircuit, despite its absence from canonical models, could be essential to basal ganglia function. DOI: http://dx.doi.org/10.7554/eLife.02397.001 PMID:24849626

Generalized Predictive Control of Dynamic Systems with Rigid-Body Modes

NASA Technical Reports Server (NTRS)

Kvaternik, Raymond G.

2013-01-01

Numerical simulations to assess the effectiveness of Generalized Predictive Control (GPC) for active control of dynamic systems having rigid-body modes are presented. GPC is a linear, time-invariant, multi-input/multi-output predictive control method that uses an ARX model to characterize the system and to design the controller. Although the method can accommodate both embedded (implicit) and explicit feedforward paths for incorporation of disturbance effects, only the case of embedded feedforward in which the disturbances are assumed to be unknown is considered here. Results from numerical simulations using mathematical models of both a free-free three-degree-of-freedom mass-spring-dashpot system and the XV-15 tiltrotor research aircraft are presented. In regulation mode operation, which calls for zero system response in the presence of disturbances, the simulations showed reductions of nearly 100%. In tracking mode operations, where the system is commanded to follow a specified path, the GPC controllers produced the desired responses, even in the presence of disturbances.

Factor analysis of auto-associative neural networks with application in speaker verification.

PubMed

Garimella, Sri; Hermansky, Hynek

2013-04-01

Auto-associative neural network (AANN) is a fully connected feed-forward neural network, trained to reconstruct its input at its output through a hidden compression layer, which has fewer numbers of nodes than the dimensionality of input. AANNs are used to model speakers in speaker verification, where a speaker-specific AANN model is obtained by adapting (or retraining) the universal background model (UBM) AANN, an AANN trained on multiple held out speakers, using corresponding speaker data. When the amount of speaker data is limited, this adaptation procedure may lead to overfitting as all the parameters of UBM-AANN are adapted. In this paper, we introduce and develop the factor analysis theory of AANNs to alleviate this problem. We hypothesize that only the weight matrix connecting the last nonlinear hidden layer and the output layer is speaker-specific, and further restrict it to a common low-dimensional subspace during adaptation. The subspace is learned using large amounts of development data, and is held fixed during adaptation. Thus, only the coordinates in a subspace, also known as i-vector, need to be estimated using speaker-specific data. The update equations are derived for learning both the common low-dimensional subspace and the i-vectors corresponding to speakers in the subspace. The resultant i-vector representation is used as a feature for the probabilistic linear discriminant analysis model. The proposed system shows promising results on the NIST-08 speaker recognition evaluation (SRE), and yields a 23% relative improvement in equal error rate over the previously proposed weighted least squares-based subspace AANNs system. The experiments on NIST-10 SRE confirm that these improvements are consistent and generalize across datasets.

Comparison of Multiple Linear Regressions and Neural Networks based QSAR models for the design of new antitubercular compounds.

PubMed

Ventura, Cristina; Latino, Diogo A R S; Martins, Filomena

2013-01-01

The performance of two QSAR methodologies, namely Multiple Linear Regressions (MLR) and Neural Networks (NN), towards the modeling and prediction of antitubercular activity was evaluated and compared. A data set of 173 potentially active compounds belonging to the hydrazide family and represented by 96 descriptors was analyzed. Models were built with Multiple Linear Regressions (MLR), single Feed-Forward Neural Networks (FFNNs), ensembles of FFNNs and Associative Neural Networks (AsNNs) using four different data sets and different types of descriptors. The predictive ability of the different techniques used were assessed and discussed on the basis of different validation criteria and results show in general a better performance of AsNNs in terms of learning ability and prediction of antitubercular behaviors when compared with all other methods. MLR have, however, the advantage of pinpointing the most relevant molecular characteristics responsible for the behavior of these compounds against Mycobacterium tuberculosis. The best results for the larger data set (94 compounds in training set and 18 in test set) were obtained with AsNNs using seven descriptors (R(2) of 0.874 and RMSE of 0.437 against R(2) of 0.845 and RMSE of 0.472 in MLRs, for test set). Counter-Propagation Neural Networks (CPNNs) were trained with the same data sets and descriptors. From the scrutiny of the weight levels in each CPNN and the information retrieved from MLRs, a rational design of potentially active compounds was attempted. Two new compounds were synthesized and tested against M. tuberculosis showing an activity close to that predicted by the majority of the models. Copyright © 2013 Elsevier Masson SAS. All rights reserved.

Handling Qualities Evaluation of Pilot Tools for Spacecraft Docking in Earth Orbit

NASA Technical Reports Server (NTRS)

Bilimoria, Karl D.; Mueller, Eric; Frost, Chad

2009-01-01

A new generation of spacecraft is now under development by NASA to replace the Space Shuttle and return astronauts to the Moon. These spacecraft will have a manual control capability for several mission tasks, and the ease and precision with which pilots can execute these tasks will have an important effect on mission risk and training costs. This paper focuses on the handling qualities of a spacecraft based on dynamics similar to that of the Crew Exploration Vehicle, during the last segment of the docking task with a space station in low Earth orbit. A previous study established that handling qualities for this task degrade significantly as the level of translation-into-rotation coupling increases. The goal of this study is to evaluate the efficacy of various pilot aids designed to mitigate the handling qualities degradation caused by this coupling. Four pilot tools were ev adluaetead:d-band box/indicator, flight-path marker, translation guidance cues, and feed-forward control. Each of these pilot tools improved handling qualities, generally with greater improvements resulting from using these tools in combination. A key result of this study is that feedforward control effectively counteracts coupling effects, providing solid Level 1 handling qualities for the spacecraft configuration evaluated.

Performance assessment of static lead-lag feedforward controllers for disturbance rejection in PID control loops.

PubMed

Yu, Zhenpeng; Wang, Jiandong

2016-09-01

This paper assesses the performance of feedforward controllers for disturbance rejection in univariate feedback plus feedforward control loops. The structures of feedback and feedforward controllers are confined to proportional-integral-derivative and static-lead-lag forms, respectively, and the effects of feedback controllers are not considered. The integral squared error (ISE) and total squared variation (TSV) are used as performance metrics. A performance index is formulated by comparing the current ISE and TSV metrics to their own lower bounds as performance benchmarks. A controller performance assessment (CPA) method is proposed to calculate the performance index from measurements. The proposed CPA method resolves two critical limitations in the existing CPA methods, in order to be consistent with industrial scenarios. Numerical and experimental examples illustrate the effectiveness of the obtained results. Copyright © 2016 ISA. Published by Elsevier Ltd. All rights reserved.

Feedforward and feedback control in apraxia of speech: effects of noise masking on vowel production.

PubMed

Maas, Edwin; Mailend, Marja-Liisa; Guenther, Frank H

2015-04-01

This study was designed to test two hypotheses about apraxia of speech (AOS) derived from the Directions Into Velocities of Articulators (DIVA) model (Guenther et al., 2006): the feedforward system deficit hypothesis and the feedback system deficit hypothesis. The authors used noise masking to minimize auditory feedback during speech. Six speakers with AOS and aphasia, 4 with aphasia without AOS, and 2 groups of speakers without impairment (younger and older adults) participated. Acoustic measures of vowel contrast, variability, and duration were analyzed. Younger, but not older, speakers without impairment showed significantly reduced vowel contrast with noise masking. Relative to older controls, the AOS group showed longer vowel durations overall (regardless of masking condition) and a greater reduction in vowel contrast under masking conditions. There were no significant differences in variability. Three of the 6 speakers with AOS demonstrated the group pattern. Speakers with aphasia without AOS did not differ from controls in contrast, duration, or variability. The greater reduction in vowel contrast with masking noise for the AOS group is consistent with the feedforward system deficit hypothesis but not with the feedback system deficit hypothesis; however, effects were small and not present in all individual speakers with AOS. Theoretical implications and alternative interpretations of these findings are discussed.

Feedforward and Feedback Control in Apraxia of Speech: Effects of Noise Masking on Vowel Production

PubMed Central

Mailend, Marja-Liisa; Guenther, Frank H.

2015-01-01

Purpose This study was designed to test two hypotheses about apraxia of speech (AOS) derived from the Directions Into Velocities of Articulators (DIVA) model (Guenther et al., 2006): the feedforward system deficit hypothesis and the feedback system deficit hypothesis. Method The authors used noise masking to minimize auditory feedback during speech. Six speakers with AOS and aphasia, 4 with aphasia without AOS, and 2 groups of speakers without impairment (younger and older adults) participated. Acoustic measures of vowel contrast, variability, and duration were analyzed. Results Younger, but not older, speakers without impairment showed significantly reduced vowel contrast with noise masking. Relative to older controls, the AOS group showed longer vowel durations overall (regardless of masking condition) and a greater reduction in vowel contrast under masking conditions. There were no significant differences in variability. Three of the 6 speakers with AOS demonstrated the group pattern. Speakers with aphasia without AOS did not differ from controls in contrast, duration, or variability. Conclusion The greater reduction in vowel contrast with masking noise for the AOS group is consistent with the feedforward system deficit hypothesis but not with the feedback system deficit hypothesis; however, effects were small and not present in all individual speakers with AOS. Theoretical implications and alternative interpretations of these findings are discussed. PMID:25565143

Multiharmonic rf feedforward system for compensation of beam loading and periodic transient effects in magnetic-alloy cavities of a proton synchrotron

NASA Astrophysics Data System (ADS)

Tamura, Fumihiko; Ohmori, Chihiro; Yamamoto, Masanobu; Yoshii, Masahito; Schnase, Alexander; Nomura, Masahiro; Toda, Makoto; Shimada, Taihei; Hasegawa, Katsushi; Hara, Keigo

2013-05-01

Beam loading compensation is a key for acceleration of a high intensity proton beam in the main ring (MR) of the Japan Proton Accelerator Research Complex (J-PARC). Magnetic alloy loaded rf cavities with a Q value of 22 are used to achieve high accelerating voltages without a tuning bias loop. The cavity is driven by a single harmonic (h=9) rf signal while the cavity frequency response also covers the neighbor harmonics (h=8,10). Therefore the wake voltage induced by the high intensity beam consists of the three harmonics, h=8,9,10. The beam loading of neighbor harmonics is the source of periodic transient effects and a possible source of coupled bunch instabilities. In the article, we analyze the wake voltage induced by the high intensity beam. We employ the rf feedforward method to compensate the beam loading of these three harmonics (h=8,9,10). The full-digital multiharmonic feedforward system was developed for the MR. We describe the system architecture and the commissioning methodology of the feedforward patterns. The commissioning of the feedforward system has been performed by using high intensity beams with 1.0×1014 proteins per pulse. The impedance seen by the beam is successfully reduced and the longitudinal oscillations due to the beam loading are reduced. By the beam loading compensation, stable high power beam operation is achieved. We also report the reduction of the momentum loss during the debunching process for the slow extraction by the feedforward.

The mechanics of state dependent neural correlations

PubMed Central

Doiron, Brent; Litwin-Kumar, Ashok; Rosenbaum, Robert; Ocker, Gabriel K.; Josić, Krešimir

2016-01-01

Simultaneous recordings from large neural populations are becoming increasingly common. An important feature of the population activity are the trial-to-trial correlated fluctuations of the spike train outputs of recorded neuron pairs. Like the firing rate of single neurons, correlated activity can be modulated by a number of factors, from changes in arousal and attentional state to learning and task engagement. However, the network mechanisms that underlie these changes are not fully understood. We review recent theoretical results that identify three separate biophysical mechanisms that modulate spike train correlations: changes in input correlations, internal fluctuations, and the transfer function of single neurons. We first examine these mechanisms in feedforward pathways, and then show how the same approach can explain the modulation of correlations in recurrent networks. Such mechanistic constraints on the modulation of population activity will be important in statistical analyses of high dimensional neural data. PMID:26906505

Detection of pseudosinusoidal epileptic seizure segments in the neonatal EEG by cascading a rule-based algorithm with a neural network.

PubMed

Karayiannis, Nicolaos B; Mukherjee, Amit; Glover, John R; Ktonas, Periklis Y; Frost, James D; Hrachovy, Richard A; Mizrahi, Eli M

2006-04-01

This paper presents an approach to detect epileptic seizure segments in the neonatal electroencephalogram (EEG) by characterizing the spectral features of the EEG waveform using a rule-based algorithm cascaded with a neural network. A rule-based algorithm screens out short segments of pseudosinusoidal EEG patterns as epileptic based on features in the power spectrum. The output of the rule-based algorithm is used to train and compare the performance of conventional feedforward neural networks and quantum neural networks. The results indicate that the trained neural networks, cascaded with the rule-based algorithm, improved the performance of the rule-based algorithm acting by itself. The evaluation of the proposed cascaded scheme for the detection of pseudosinusoidal seizure segments reveals its potential as a building block of the automated seizure detection system under development.

Neural Network Compensation for Frequency Cross-Talk in Laser Interferometry

NASA Astrophysics Data System (ADS)

Lee, Wooram; Heo, Gunhaeng; You, Kwanho

The heterodyne laser interferometer acts as an ultra-precise measurement apparatus in semiconductor manufacture. However the periodical nonlinearity property caused from frequency cross-talk is an obstacle to improve the high measurement accuracy in nanometer scale. In order to minimize the nonlinearity error of the heterodyne interferometer, we propose a frequency cross-talk compensation algorithm using an artificial intelligence method. The feedforward neural network trained by back-propagation compensates the nonlinearity error and regulates to minimize the difference with the reference signal. With some experimental results, the improved accuracy is proved through comparison with the position value from a capacitive displacement sensor.

Ship localization in Santa Barbara Channel using machine learning classifiers.

PubMed

Niu, Haiqiang; Ozanich, Emma; Gerstoft, Peter

2017-11-01

Machine learning classifiers are shown to outperform conventional matched field processing for a deep water (600 m depth) ocean acoustic-based ship range estimation problem in the Santa Barbara Channel Experiment when limited environmental information is known. Recordings of three different ships of opportunity on a vertical array were used as training and test data for the feed-forward neural network and support vector machine classifiers, demonstrating the feasibility of machine learning methods to locate unseen sources. The classifiers perform well up to 10 km range whereas the conventional matched field processing fails at about 4 km range without accurate environmental information.

Compensation of significant parametric uncertainties using sliding mode online learning

NASA Astrophysics Data System (ADS)

Schnetter, Philipp; Kruger, Thomas

An augmented nonlinear inverse dynamics (NID) flight control strategy using sliding mode online learning for a small unmanned aircraft system (UAS) is presented. Because parameter identification for this class of aircraft often is not valid throughout the complete flight envelope, aerodynamic parameters used for model based control strategies may show significant deviations. For the concept of feedback linearization this leads to inversion errors that in combination with the distinctive susceptibility of small UAS towards atmospheric turbulence pose a demanding control task for these systems. In this work an adaptive flight control strategy using feedforward neural networks for counteracting such nonlinear effects is augmented with the concept of sliding mode control (SMC). SMC-learning is derived from variable structure theory. It considers a neural network and its training as a control problem. It is shown that by the dynamic calculation of the learning rates, stability can be guaranteed and thus increase the robustness against external disturbances and system failures. With the resulting higher speed of convergence a wide range of simultaneously occurring disturbances can be compensated. The SMC-based flight controller is tested and compared to the standard gradient descent (GD) backpropagation algorithm under the influence of significant model uncertainties and system failures.

A Dual-Stream Neuroanatomy of Singing

PubMed Central

Loui, Psyche

2015-01-01

Singing requires effortless and efficient use of auditory and motor systems that center around the perception and production of the human voice. Although perception and production are usually tightly coupled functions, occasional mismatches between the two systems inform us of dissociable pathways in the brain systems that enable singing. Here I review the literature on perception and production in the auditory modality, and propose a dual-stream neuroanatomical model that subserves singing. I will discuss studies surrounding the neural functions of feedforward, feedback, and efference systems that control vocal monitoring, as well as the white matter pathways that connect frontal and temporal regions that are involved in perception and production. I will also consider disruptions of the perception-production network that are evident in tone-deaf individuals and poor pitch singers. Finally, by comparing expert singers against other musicians and nonmusicians, I will evaluate the possibility that singing training might offer rehabilitation from these disruptions through neuroplasticity of the perception-production network. Taken together, the best available evidence supports a model of dorsal and ventral pathways in auditory-motor integration that enables singing and is shared with language, music, speech, and human interactions in the auditory environment. PMID:26120242

A Computational Analysis of Neural Mechanisms Underlying the Maturation of Multisensory Speech Integration in Neurotypical Children and Those on the Autism Spectrum

PubMed Central

Cuppini, Cristiano; Ursino, Mauro; Magosso, Elisa; Ross, Lars A.; Foxe, John J.; Molholm, Sophie

2017-01-01

Failure to appropriately develop multisensory integration (MSI) of audiovisual speech may affect a child's ability to attain optimal communication. Studies have shown protracted development of MSI into late-childhood and identified deficits in MSI in children with an autism spectrum disorder (ASD). Currently, the neural basis of acquisition of this ability is not well understood. Here, we developed a computational model informed by neurophysiology to analyze possible mechanisms underlying MSI maturation, and its delayed development in ASD. The model posits that strengthening of feedforward and cross-sensory connections, responsible for the alignment of auditory and visual speech sound representations in posterior superior temporal gyrus/sulcus, can explain behavioral data on the acquisition of MSI. This was simulated by a training phase during which the network was exposed to unisensory and multisensory stimuli, and projections were crafted by Hebbian rules of potentiation and depression. In its mature architecture, the network also reproduced the well-known multisensory McGurk speech effect. Deficits in audiovisual speech perception in ASD were well accounted for by fewer multisensory exposures, compatible with a lack of attention, but not by reduced synaptic connectivity or synaptic plasticity. PMID:29163099

A Dual-Stream Neuroanatomy of Singing.

PubMed

Loui, Psyche

2015-02-01

Singing requires effortless and efficient use of auditory and motor systems that center around the perception and production of the human voice. Although perception and production are usually tightly coupled functions, occasional mismatches between the two systems inform us of dissociable pathways in the brain systems that enable singing. Here I review the literature on perception and production in the auditory modality, and propose a dual-stream neuroanatomical model that subserves singing. I will discuss studies surrounding the neural functions of feedforward, feedback, and efference systems that control vocal monitoring, as well as the white matter pathways that connect frontal and temporal regions that are involved in perception and production. I will also consider disruptions of the perception-production network that are evident in tone-deaf individuals and poor pitch singers. Finally, by comparing expert singers against other musicians and nonmusicians, I will evaluate the possibility that singing training might offer rehabilitation from these disruptions through neuroplasticity of the perception-production network. Taken together, the best available evidence supports a model of dorsal and ventral pathways in auditory-motor integration that enables singing and is shared with language, music, speech, and human interactions in the auditory environment.

Optimal feed-forward compensation for PWM dc/dc converters with 'linear' and 'quadratic' conversion ratio

NASA Astrophysics Data System (ADS)

Calderone, Luigi; Pinola, Licia; Varoli, Vincenzo

1992-04-01

The paper describes an analytical procedure to optimize the feed-forward compensation for any PWM dc/dc converters. The aims of achieving zero dc audiosusceptibility was found to be possible for the buck, buck-boost, Cuk, and SEPIC cells; for the boost converter, however, only nonoptimal compensation is feasible. Rules for the design of PWM controllers and procedures for the evaluation of the hardware-introduced errors are discussed. A PWM controller implementing the optimal feed-forward compensation for buck-boost, Cuk, and SEPIC cells is described and fully experimentally characterized.

Hypocortisolemic clamp unmasks jointly feedforward- and feedback-dependent control of overnight ACTH secretion

PubMed Central

Iranmanesh, Ali; Veldhuis, Johannes D

2009-01-01

Background ACTH secretion is under hypothalamic stimulatory (feedforward) and adrenal inhibitory (feedback) control. Hypothesis Assessment of overnight ACTH secretion during a hypocortisolemic clamp will permit the estimation of changing feedforward and feedback. Subjects Seven healthy men. Interventions An oral dose of placebo (PLAC), metyrapone (METY, 3 g), or ketoconazole (KTCZ, 1.2 g) was given at midnight (MN) to block glucocorticoid synthesis. Plasma ACTH was sampled every 10 min (MN to 0800 h). Analysis Variable-waveform deconvolution analysis of ACTH secretion and approximate entropy (ApEn) analysis of pattern regularity. Results Compared with PLAC, administration of METY and KTCZ reduced morning cortisol concentrations by ≥77 and 54% respectively (P<0.001). Hypocortisolemia elevated pulsatile ACTH secretion by 8.2- (METY) and 5.3-fold (KTCZ; both P<0.001). Basal ACTH secretion rose by 3.4-fold under METY-induced cortisol depletion (P = 0.020). ACTH secretory-burst shape and half-life were stable. ApEn of ACTH release declined overnight (P = 0.021) and with the drug (P = 0.001), denoting enhanced feedforward coordination. Conclusion The combined data predict overnight amplification and coordination of hypothalamic feedforward drive onto ACTH release. Therefore, disruption of either mechanism might contribute to clinical pathophysiology, such as late-day elevations of cortisol output in fasting, alcoholism, depression, or aging. PMID:18713842

A stochastic optimal feedforward and feedback control methodology for superagility

NASA Technical Reports Server (NTRS)

Halyo, Nesim; Direskeneli, Haldun; Taylor, Deborah B.

1992-01-01

A new control design methodology is developed: Stochastic Optimal Feedforward and Feedback Technology (SOFFT). Traditional design techniques optimize a single cost function (which expresses the design objectives) to obtain both the feedforward and feedback control laws. This approach places conflicting demands on the control law such as fast tracking versus noise atttenuation/disturbance rejection. In the SOFFT approach, two cost functions are defined. The feedforward control law is designed to optimize one cost function, the feedback optimizes the other. By separating the design objectives and decoupling the feedforward and feedback design processes, both objectives can be achieved fully. A new measure of command tracking performance, Z-plots, is also developed. By analyzing these plots at off-nominal conditions, the sensitivity or robustness of the system in tracking commands can be predicted. Z-plots provide an important tool for designing robust control systems. The Variable-Gain SOFFT methodology was used to design a flight control system for the F/A-18 aircraft. It is shown that SOFFT can be used to expand the operating regime and provide greater performance (flying/handling qualities) throughout the extended flight regime. This work was performed under the NASA SBIR program. ICS plans to market the software developed as a new module in its commercial CACSD software package: ACET.