Many current computational models that aim to simulate cortical and hippocampal modules of the brain depend on artificial neural networks. However, such classical or even deep neural networks are very slow, sometimes ...
详细信息
Many current computational models that aim to simulate cortical and hippocampal modules of the brain depend on artificial neural networks. However, such classical or even deep neural networks are very slow, sometimes taking thousands of trials to obtain the final response with a considerable amount of error. The need for a large number of trials at learning and the inaccurate output responses are due to the complexity of the input cue and the biological processes being simulated. This article proposes a computational model for an intact and a lesioned cortico-hippocampal system using quantum-inspired neural networks. This cortico-hippocampal computational quantum-inspired (CHCQI) model simulates cortical and hippocampal modules by using adaptively updated neural networks entangled with quantum circuits. The proposed model is used to simulate various classical conditioning tasks related to biological processes. The output of the simulated tasks yielded the desired responses quickly and efficiently compared with other computational models, including the recently published Green model.
Recently, Ethernet Passive Optical Network (EPON) is a significant solution in the access layer of network that can provide large amount of bandwidth for different kind of traffic. In this paper, we propose a dynamic ...
详细信息
Recently, Ethernet Passive Optical Network (EPON) is a significant solution in the access layer of network that can provide large amount of bandwidth for different kind of traffic. In this paper, we propose a dynamic bandwidth allocation scheme based on the trust based VCG-Kelly mechanism. To enhance the EPON system performance, the proposed scheme can take into account the measure of each ONU's payoff variation succeeding at a given bandwidth allocation. By considering the current outcome, ONUs adaptively select their bid strategies and the OLT dynamically allocates bandwidth to reach a certain desired EPON outcome. In addition, we also consider the network security situation, which reflects what is happening in the EPON network including both the offense and defense behaviors. During this iterative learning approach, the EPON system can converge to a stable network state. Evidences from simulations demonstrate that the proposed scheme outperforms the existing schemes.
Capacity control in perceptron decision trees is typically performed by controlling their size. We prove that other quantities can be as relevant to reduce their flexibility and combat overfitting. In particular, we p...
详细信息
Capacity control in perceptron decision trees is typically performed by controlling their size. We prove that other quantities can be as relevant to reduce their flexibility and combat overfitting. In particular, we provide an upper bound on the generalization error which depends both on the size of the tree and on the margin of the decision nodes. So enlarging the margin in perceptron decision trees will reduce the upper bound on generalization error. Based on this analysis, we introduce three new algorithms, which can induce large margin perceptron decision trees. To assess the effect of the large margin bias, OC1 (Journal of Artificial Intelligence Research, 1994, 2, 1-32.) of Murthy, Kasif and Salzberg, a well-known system for inducing perceptron decision trees, is used as the baseline algorithm. An extensive experimental study on real world data showed that all three new algorithms perform better or at least not significantly worse than OC1 on almost every dataset with only one exception. OC1 performed worse than the best margin-based method on every dataset.
Nonlinear equalisers based on minimum BER are proposed for the equalisation of nonlinear time-varying channels. To train the equalisers online, a sliding-window-based hybrid quasi-Newton algorithm is proposed. Switchi...
详细信息
Nonlinear equalisers based on minimum BER are proposed for the equalisation of nonlinear time-varying channels. To train the equalisers online, a sliding-window-based hybrid quasi-Newton algorithm is proposed. Switching between sliding-window stochastic gradient algorithm and sliding-window quasi-Newton algorithm makes the new algorithm significantly stabler with a fast convergence rate. Results from extensive simulation tests show that performance of nonlinear equalisers based on minimum BER is better than the equaliser based on minimum mean square error. The proposed algorithm demonstrates high efficiency as well. Copyright (c) 2013 John Wiley & Sons, Ltd.
In order to make device-to-device (D2D) content sharing give full play to its advantage of improving local area services, one of the important issues is to decide the channels that D2D pairs occupy. Most existing work...
详细信息
In order to make device-to-device (D2D) content sharing give full play to its advantage of improving local area services, one of the important issues is to decide the channels that D2D pairs occupy. Most existing works study this issue in static environment, and ignore the guidance for D2D pairs to select the channel adaptively. In this paper, we investigate this issue in dynamic environment where D2D pairs' activeness and wireless channel are dynamic. Specifically, we propose a pricing-based approach to guide D2D pairs to select different channels according to the spectrum resource states adaptively. Then, we formulate the pricing-based channel selection problem as an expected global price-to-performance ratio minimum problem. In order to solve it in a tractable manner, we make an approximately equivalent transformation to it. After that, we model the transformed problem as a stochastic game and prove it to be an exact potential game, which has at least one pure strategy Nash Equilibrium (NE) point. In order to reach the pure strategy NE points in dynamic environment, we design a channel selection learning algorithm based on stochastic learning automata, which only requires little information exchange. Simulation results show that our proposed algorithm outperforms other benchmark algorithms.
This paper discusses neural network-based strategy for reducing the existing errors of fiber-optic gyroscope (FOG), A series-single-layer neural network, which is composed of two single-layer networks in series, is pr...
详细信息
This paper discusses neural network-based strategy for reducing the existing errors of fiber-optic gyroscope (FOG), A series-single-layer neural network, which is composed of two single-layer networks in series, is presented for eliminating random noises. This network has simpler architecture, faster learning speed, and better performance compared to conventional backpropagation (BP) networks. Accordingly, after considering the characteristics of the power law noise in FOG, an advanced learning algorithm is proposed by using the increments of errors in energy function. Furthermore, a radial basis function (RBF) neural network-based method is also posed to evaluate and compensate the temperature drift of FOG. The orthogonal least squares (OLS) algorithm is applied due to its simplicity, high accuracy, and fast learning speed. The simulation results show that the series-single-layer network (SSLN) with the advanced learning algorithm provides a fast and effective way for eliminating different random noises including stable and unstable noises existing in FOG, and the RBF network-based method offers a powerful and successful procedure for evaluating and compensating the temperature drift.
An intelligent power factor correction approach based on artificial neural networks (ANN) is introduced. Four learning algorithms, backpropagation (BP), delta-bar-delta (DBD), extended delta-bar-delta (EDBD) and direc...
详细信息
An intelligent power factor correction approach based on artificial neural networks (ANN) is introduced. Four learning algorithms, backpropagation (BP), delta-bar-delta (DBD), extended delta-bar-delta (EDBD) and directed random search (DRS), were used to train the ANNs. The best test results obtained from the ANN compensators trained with the four learning algorithms were first achieved. The parameters belonging to each neural compensator obtained from an off-line training were then inserted into a microcontroller for on-line usage. The results have shown that the selected intelligent compensators developed in this work might overcome the problems occured in the literature providing accurate, simple and low-cost solution for compensation. (C) 2008 Elsevier B.V. All rights reserved.
This article deals with the development of four modified radial basis function neural network (RBFNN) models. The corresponding learning algorithms associated with the updating of internal parameters of the models are...
详细信息
This article deals with the development of four modified radial basis function neural network (RBFNN) models. The corresponding learning algorithms associated with the updating of internal parameters of the models are derived. The conventional inputs are used in the first and second modified RBFNN models (models 3 and 4) whereas exponential nonlinear inputs are used in the fifth and sixth RBFNN models to provide additional nonlinearity for achieving a better solution of nonlinear classification, and direct and inverse modeling problems. To assess and compare the performance potentiality of the proposed four new RBFNN models, one classification problem, one direct modeling problem, and one inverse modeling problem are solved through computer simulation-based experiments. For comparison and to assign the performance rank of each of the four modified RBFNN models, two conventional and commonly used RBFNN models (models 1 and 2) are also simulated. To access the performance of different models during the training phase of Examples 1 and 2, the root mean-square error (RMSE) value, mean absolute deviation (MAD), and the number of iterations required to achieve convergence are obtained. For the third example, only the first two performance measures are found. During the testing or validation phase, the output responses of the different models of Example 2 are compared with the desired response analysis. For Example 3, the bit-error rate (BER) plots are compared. The observation of all the results demonstrates consistent ranks of all models in the case of all three examples. It is, in general, found that the ranks of the models 1-6 are 6, 4, 3, 2, 5, and 1, respectively. In essence, in terms of all performance measures, model M-6 with an exponential version of inputs with weights on both layers occupies the first position whereas model M-4 with conventional inputs as the second position.
Using the experimental data obtained from hot compression tests in the temperature range 800-1200 degrees C, strain range 0.05-0.90, and strain rate range 0.01-50 s(-1), an artificial neural network (ANN) model is dev...
详细信息
Using the experimental data obtained from hot compression tests in the temperature range 800-1200 degrees C, strain range 0.05-0.90, and strain rate range 0.01-50 s(-1), an artificial neural network (ANN) model is developed to predict the hot deformation behavior of the ultrahigh strength steel of Aermet100. The inputs of the neural network are strain, strain rate and temperature, whereas flow stress is the output. The developed feed-forward back-propagation ANN model is trained with Levenberg-Marquardt learning algorithm. The performance of the ANN model is evaluated using a wide variety of standard statistical indices. Results show that the ANN model can efficiently and accurately predict hot deformation behavior of Aermet100. Finally the extrapolation ability and noise sensitivity of the ANN model are also investigated. It is found that the extrapolation ability is very high in the proximity of the training domain, and the noise tolerance ability very robust. (C) 2010 Elsevier B.V. All rights reserved.
暂无评论