Enhancing weak signal transmission through a feedforward network

The ability to transmit and amplify weak signals is fundamental to signal processing of artificial devices in engineering. Using a multilayer feedforward network of coupled double-well oscillators as well as Fitzhugh-Nagumo oscillators, we here investigate the conditions under which a weak signal received by the first layer can be transmitted through the network with or without amplitude attenuation. We find that the coupling strength and the nodes' states of the first layer act as two-state switches, which determine whether the transmission is significantly enhanced or exponentially decreased. We hope this finding is useful for designing artificial signal amplifiers.

The sample complexity of pattern classification with neural networks: the size of the weights is more important than the size of the network

Sample complexity results from computational learning theory, when applied to neural network learning for pattern classification problems, suggest that for good generalization performance the number of training examples should grow at least linearly with the number of adjustable parameters in the network. Results in this paper show that if a large neural network is used for a pattern classification problem and the learning algorithm finds a network with small weights that has small squared error on the training patterns, then the generalization performance depends on the size of the weights rather than the number of weights. For example, consider a two-layer feedforward network of sigmoid units, in which the sum of the magnitudes of the weights associated with each unit is bounded by A and the input dimension is n. We show that the misclassification probability is no more than a certain error estimate (that is related to squared error on the training set) plus A3 √((log n)/m) (ignoring log A and log m factors), where m is the number of training patterns. This may explain the generalization performance of neural networks, particularly when the number of training examples is considerably smaller than the number of weights. It also supports heuristics (such as weight decay and early stopping) that attempt to keep the weights small during training. The proof techniques appear to be useful for the analysis of other pattern classifiers: when the input domain is a totally bounded metric space, we use the same approach to give upper bounds on misclassification probability for classifiers with decision boundaries that are far from the training examples.

Use of a Neural Network to Predict Strength and Optimum Compositions of Natural Alumina-Silica-Based Geopolymers

In order to predict compressive strength of geopolymers prepared from alumina-silica natural products, based on the effect of Al 2 O 3 /SiO 2, Na 2 O/Al 2 O 3, Na 2 O/H 2 O, and Na/[Na+K], more than 50 pieces of data were gathered from the literature. The data was utilized to train and test a multilayer artificial neural network (ANN). Therefore a multilayer feedforward network was designed with chemical compositions of alumina silicate and alkali activators as inputs and compressive strength as output. In this study, a feedforward network with various numbers of hidden layers and neurons were tested to select the optimum network architecture. The developed three-layer neural network simulator model used the feedforward back propagation architecture, demonstrated its ability in training the given input/output patterns. The cross-validation data was used to show the validity and high prediction accuracy of the network. This leads to the optimum chemical composition and the best paste can be made from activated alumina-silica natural products using alkaline hydroxide, and alkaline silicate. The research results are in agreement with mechanism of geopolymerization.


Read More: http://ascelibrary.org/doi/abs/10.1061/(ASCE)MT.1943-5533.0000829

Learning the global maximum with parameterized learning automata

A feedforward network composed of units of teams of parameterized learning automata is considered as a model of a reinforcement teaming system. The internal state vector of each learning automaton is updated using an algorithm consisting of a gradient following term and a random perturbation term. It is shown that the algorithm weakly converges to a solution of the Langevin equation implying that the algorithm globally maximizes an appropriate function. The algorithm is decentralized, and the units do not have any information exchange during updating. Simulation results on common payoff games and pattern recognition problems show that reasonable rates of convergence can be obtained.

Neural networks for time-varying data

This paper reviews some basic issues and methods involved in using neural networks to respond in a desired fashion to a temporally-varying environment. Some popular network models and training methods are introduced. A speech recognition example is then used to illustrate the central difficulty of temporal data processing: learning to notice and remember relevant contextual information. Feedforward network methods are applicable to cases where this problem is not severe. The application of these methods are explained and applications are discussed in the areas of pure mathematics, chemical and physical systems, and economic systems. A more powerful but less practical algorithm for temporal problems, the moving targets algorithm, is sketched and discussed. For completeness, a few remarks are made on reinforcement learning.

Hypothalamic feedforward inhibition of thalamocortical network controls arousal and consciousness.

During non-rapid eye movement (NREM) sleep, synchronous synaptic activity in the thalamocortical network generates predominantly low-frequency oscillations (<4 Hz) that are modulated by inhibitory inputs from the thalamic reticular nucleus (TRN). Whether TRN cells integrate sleep-wake signals from subcortical circuits remains unclear. We found that GABA neurons from the lateral hypothalamus (LHGABA) exert a strong inhibitory control over TRN GABA neurons (TRNGABA). We found that optogenetic activation of this circuit recapitulated state-dependent changes of TRN neuron activity in behaving mice and induced rapid arousal during NREM, but not REM, sleep. During deep anesthesia, activation of this circuit induced sustained cortical arousal. In contrast, optogenetic silencing of LHGABA-TRNGABA transmission increased the duration of NREM sleep and amplitude of delta (1-4 Hz) oscillations. Collectively, these results demonstrate that TRN cells integrate subcortical arousal inputs selectively during NREM sleep and may participate in sleep intensity.

Analyzing MSI rules for the USA extracted from a feedforward neural network

This paper introduces a mechanism for generating a series of rules that characterize the money price relationship for the USA, defined as the relationship between the rate of growth of the money supply and inflation. Monetary component data is used to train a selection of candidate feedforward neural networks. The selected network is mined for rules, expressed in human-readable and machine-executable form. The rule and network accuracy are compared, and expert commentary is made on the readability and reliability of the extracted rule set. The ultimate goal of this research is to produce rules that meaningfully and accurately describe inflation in terms of the monetary component dataset.

A fuzzy neural network approach for contractor pre-qualification

Nonlinearity, uncertainty and subjectivity are the three predominant characteristics of contractors prequalification which cause the process more of an art than a scientific evaluation. A fuzzy neural network (FNN) model, amalgamating both the fuzzy set and neural network theories, has been developed aiming to improve the objectiveness of contractor prequalification. Through the FNN theory, the fuzzy rules as used by the prequalifiers can be identified and the corresponding membership functions can be transformed. Eighty-five cases with detailed decision criteria and rules for prequalifying Hong Kong civil engineering contractors were collected. These cases were used for training (calibrating) and testing the FNN model. The performance of the FNN model was compared with the original results produced by the prequalifiers and those generated by the general feedforward neural network (GFNN, i.e. a crisp neural network) approach. Contractor’s ranking orders, the model efficiency (R2) and the mean absolute percentage error (MAPE) were examined during the testing phase. These results indicate the applicability of the neural network approach for contractor prequalification and the benefits of the FNN model over the GFNN model. The FNN is a practical approach for modelling contractor prequalification.

Filter design using radial basis function neural network and genetic algorithm for improved operational health monitoring

The problem of denoising damage indicator signals for improved operational health monitoring of systems is addressed by applying soft computing methods to design filters. Since measured data in operational settings is contaminated with noise and outliers, pattern recognition algorithms for fault detection and isolation can give false alarms. A direct approach to improving the fault detection and isolation is to remove noise and outliers from time series of measured data or damage indicators before performing fault detection and isolation. Many popular signal-processing approaches do not work well with damage indicator signals, which can contain sudden changes due to abrupt faults and non-Gaussian outliers. Signal-processing algorithms based on radial basis function (RBF) neural network and weighted recursive median (WRM) filters are explored for denoising simulated time series. The RBF neural network filter is developed using a K-means clustering algorithm and is much less computationally expensive to develop than feedforward neural networks trained using backpropagation. The nonlinear multimodal integer-programming problem of selecting optimal integer weights of the WRM filter is solved using genetic algorithm. Numerical results are obtained for helicopter rotor structural damage indicators based on simulated frequencies. Test signals consider low order polynomial growth of damage indicators with time to simulate gradual or incipient faults and step changes in the signal to simulate abrupt faults. Noise and outliers are added to the test signals. The WRM and RBF filters result in a noise reduction of 54 - 71 and 59 - 73% for the test signals considered in this study, respectively. Their performance is much better than the moving average FIR filter, which causes significant feature distortion and has poor outlier removal capabilities and shows the potential of soft computing methods for specific signal-processing applications.

Filter design using radial basis function neural network and genetic algorithm for improved operational health monitoring

The problem of denoising damage indicator signals for improved operational health monitoring of systems is addressed by applying soft computing methods to design filters. Since measured data in operational settings is contaminated with noise and outliers, pattern recognition algorithms for fault detection and isolation can give false alarms. A direct approach to improving the fault detection and isolation is to remove noise and outliers from time series of measured data or damage indicators before performing fault detection and isolation. Many popular signal-processing approaches do not work well with damage indicator signals, which can contain sudden changes due to abrupt faults and non-Gaussian outliers. Signal-processing algorithms based on radial basis function (RBF) neural network and weighted recursive median (WRM) filters are explored for denoising simulated time series. The RBF neural network filter is developed using a K-means clustering algorithm and is much less computationally expensive to develop than feedforward neural networks trained using backpropagation. The nonlinear multimodal integer-programming problem of selecting optimal integer weights of the WRM filter is solved using genetic algorithm. Numerical results are obtained for helicopter rotor structural damage indicators based on simulated frequencies. Test signals consider low order polynomial growth of damage indicators with time to simulate gradual or incipient faults and step changes in the signal to simulate abrupt faults. Noise and outliers are added to the test signals. The WRM and RBF filters result in a noise reduction of 54 - 71 and 59 - 73% for the test signals considered in this study, respectively. Their performance is much better than the moving average FIR filter, which causes significant feature distortion and has poor outlier removal capabilities and shows the potential of soft computing methods for specific signal-processing applications. (C) 2005 Elsevier B. V. All rights reserved.

Control of a Duct Flow Network for Wind Display in a Virtual Environment

Control of flow in duct networks has a myriad of applications ranging from heating, ventilation, and air-conditioning to blood flow networks. The system considered here provides vent velocity inputs to a novel 3-D wind display device called the TreadPort Active Wind Tunnel. An error-based robust decentralized sliding-mode control method with nominal feedforward terms is developed for individual ducts while considering cross coupling between ducts and model uncertainty as external disturbances in the output. This approach is important due to limited measurements, geometric complexities, and turbulent flow conditions. Methods for resolving challenges such as turbulence, electrical noise, valve actuator design, and sensor placement are presented. The efficacy of the controller and the importance of feedforward terms are demonstrated with simulations based upon an experimentally validated lumped parameter model and experiments on the physical system. Results show significant improvement over traditional control methods and validate prior assertions regarding the importance of decentralized control in practice.

Synconset Waves and Chains: Spiking Onsets in Synchronous Populations Predict and Are Predicted by Network Structure

Synfire waves are propagating spike packets in synfire chains, which are feedforward chains embedded in random networks. Although synfire waves have proved to be effective quantification for network activity with clear relations to network structure, their utilities are largely limited to feedforward networks with low background activity. To overcome these shortcomings, we describe a novel generalisation of synfire waves, and define `synconset wave' as a cascade of first spikes within a synchronisation event. Synconset waves would occur in `synconset chains', which are feedforward chains embedded in possibly heavily recurrent networks with heavy background activity. We probed the utility of synconset waves using simulation of single compartment neuron network models with biophysically realistic conductances, and demonstrated that the spread of synconset waves directly follows from the network connectivity matrix and is modulated by top-down inputs and the resultant oscillations. Such synconset profiles lend intuitive insights into network organisation in terms of connection probabilities between various network regions rather than an adjacency matrix. To test this intuition, we develop a Bayesian likelihood function that quantifies the probability that an observed synfire wave was caused by a given network. Further, we demonstrate it's utility in the inverse problem of identifying the network that caused a given synfire wave. This method was effective even in highly subsampled networks where only a small subset of neurons were accessible, thus showing it's utility in experimental estimation of connectomes in real neuronal-networks. Together, we propose synconset chains/waves as an effective framework for understanding the impact of network structure on function, and as a step towards developing physiology-driven network identification methods. Finally, as synconset chains extend the utilities of synfire chains to arbitrary networks, we suggest utilities of our framework to several aspects of network physiology including cell assemblies, population codes, and oscillatory synchrony.

A What-and-Where Neural Network for Invariant Image Preprocessing

A feedforward neural network for invariant image preprocessing is proposed that represents the position1 orientation and size of an image figure (where it is) in a multiplexed spatial map. This map is used to generate an invariant representation of the figure that is insensitive to position1 orientation, and size for purposes of pattern recognition (what it is). A multiscale array of oriented filters followed by competition between orientations and scales is used to define the Where filter.

Neural network controlled three-phase four-wire shunt active powerfilter

A three-phase four-wire shunt active power filter for harmonic mitigation and reactive power compensation in power systems supplying nonlinear loads is presented. Three adaptive linear neurons are used to tackle the desired three-phase filter current templates. Another feedforward three-layer neural network is adopted to control the output filter compensating currents online. This is accomplished by producing the appropriate switching patterns of the converter's legs IGBTs. Adequate tracking of the filter current references is obtained by this method. The active filter injects the current required to compensate for the harmonic and reactive components of the line currents, Simulation results of the proposed active filter indicate a remarkable improvement in the source current waveforms. This is reflected in the enhancement of the unified power quality index defined. Also, the filter has exhibited quite a high dynamic response for step variations in the load current, assuring its potential for real-time applications

Improving neural network training solutions using regularisation

This paper describes the application of regularisation to the training of feedforward neural networks, as a means of improving the quality of solutions obtained. The basic principles of regularisation theory are outlined for both linear and nonlinear training and then extended to cover a new hybrid training algorithm for feedforward neural networks recently proposed by the authors. The concept of functional regularisation is also introduced and discussed in relation to MLP and RBF networks. The tendency for the hybrid training algorithm and many linear optimisation strategies to generate large magnitude weight solutions when applied to ill-conditioned neural paradigms is illustrated graphically and reasoned analytically. While such weight solutions do not generally result in poor fits, it is argued that they could be subject to numerical instability and are therefore undesirable. Using an illustrative example it is shown that, as well as being beneficial from a generalisation perspective, regularisation also provides a means for controlling the magnitude of solutions. (C) 2001 Elsevier Science B.V. All rights reserved.