检索结果-内蒙古大学图书馆

IEEE International Conference on Systems, Man and Cybernetics

作者： Kwang Baek Kim Myung Ho Kang Eui Young Cha Department of Computer Science Pusan Women's University busan South Korea

ISBN: (纸本)0780340531

We consider a fuzzy multilayer neural network with a single hidden layer. A conventional backpropagation learning algorithm used widely in multilayer neural network has a possibility of local minima due to the inadequate initial weights and the insufficient number of hidden nodes. So we propose a fuzzy self-organized backpropagation algorithm that self-generates hidden nodes by a compound method of supervised learning and unsupervised learning. From the input layer to the hidden layer, a modified fuzzy ART is used to produce nodes. A winner-take-all method is adopted to the connection weight adaption, so that stored pattern for some pattern becomes updated. In simulation results, our method reduces the possibility of local minima and improves learning speed and paralysis than conventional error backpropagation.

关键词： Fuzzy systems Nervous system Multi-layer neural network Biological neural networks Neural networks backpropagation algorithms Fuzzy neural networks Supervised learning Unsupervised learning Subspace constraints

来源：评论

学校读者我要写书评

暂无评论

Incorporation of fuzzy classification properties into backpropagation learning algorithm

Incorporation of fuzzy classification properties into backpr...

引用

IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)

作者： M. Sarkar B. Yegnanarayana Department of Computer Science & Engineering Indian Institute of Technology Chennai India

Most of the real life classification problems have ill defined, imprecise or fuzzy class boundaries. Feedforward neural networks with conventional backpropagation learning algorithm are not tailored to these kinds of classification problems. Hence, in this paper, feedforward neural networks, that use fuzzy objective functions in the backpropagation learning algorithm, are investigated. A learning algorithm is proposed that minimizes an error term, which takes care of fuzziness in classification from the point of view of possibilistic approach. Since the proposed algorithm has possibilistic classification ability, it can encompass different backpropagation learning algorithms based on crisp and constrained fuzzy classification. The efficacy of the proposed scheme is demonstrated on a vowel classification problem.

关键词： backpropagation algorithms Neural networks Feedforward neural networks Fuzzy neural networks Mean square error methods Entropy Computer science Fuzzy sets Marine vehicles

来源：评论

学校读者我要写书评

暂无评论

Decision feedback equalization using complex backpropagation algorithm

Decision feedback equalization using complex backpropagation...

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Ching-Haur Chang Sammy Siu Che-Ho Wei Chunghwa Telecom Training Institute Taichung Branch Taichung Taiwan Chunghwa Telecom Laboratories Chungli Taiwan Institute of Electronics National Chiao Tung University Hsinchu Taiwan

A decision feedback equalizer (DFE) using complex backpropagation (CBP) algorithm is presented in this paper. In passband multilevel signaling systems, learning algorithms with complex-valued parameters are generally required in adaptive equalizers. Conventional approaches using complex least mean square (CLMS) algorithm cannot be satisfactory due to their lack of capability in partitioning the signal space. A new approach using CBP is used to equip a DFE with better performance. Computer simulations show significant performance improvement compared to the conventional one.

关键词： Decision feedback equalizers backpropagation algorithms Neurons Quadrature phase shift keying Neurofeedback Delay estimation Frequency Quadrature amplitude modulation Communication system signaling Signal mapping

来源：评论

学校读者我要写书评

暂无评论

Synaptic delay based artificial neural networks and discrete time backpropagation applied to QRS complex detection

Synaptic delay based artificial neural networks and discrete...

引用

International Conference on Neural Networks

作者： R.J. Duro J. Santos Dpto. Ingeniería Industrial Universidad da la Coruňa La Coruna Spain Dpto. Computación Universidad de La Coruna La Coruna Spain

In this paper we make use of an extension of the backpropagation algorithm to discrete time feedforward networks that include internal time delays in the synapses. The structure of the network is similar to the one presented by Day-Davenport (1993), that is, in addition to the weights of the synaptic connections, we model their length through a parameter that indicates the delay a discrete event suffers when going from the origin neuron to the target neuron through a synaptic connection. Like the weights, these delays are also trainable, and a training algorithm can be obtained that is almost as simple as the backpropagation algorithm, and which is really an extension of it. We present an application of these networks to the task of identifying normal QRS and ventricular QRS complexes in an ECG signal with the network receiving the signal sequentially, that is, no windowing or segmentation is applied.

关键词： Delay effects Artificial neural networks backpropagation algorithms Signal processing algorithms Signal processing Neurons Computer networks Electronic mail Feeds Electrocardiography

来源：评论

学校读者我要写书评

暂无评论

Parallel implementation of backpropagation neural network on a heterogeneous ring processor topology

Parallel implementation of backpropagation neural network on...

引用

International Conference on Neural Networks

作者： Shou King Foo P. Saratchandran N. Sundararajan School of Electrical & Electronic Engineering Nanyang Technological University Singapore

This paper analyzes the parallel mapping of the backpropagation learning algorithm, onto a heterogeneous multiprocessor ring architecture. Training set parallelism is used as the parallelizing paradigm. A mathematical model is developed in order to obtain an expression for the training time per epoch. This model is then used to find the optimal mapping that minimizes the training time per epoch. It is shown that the optimal mapping results in a mixed integer programming (MIP) problem which is NP-complete. To solve this problem, the genetic algorithmic approach is used and the optimal distribution of training patterns are obtained. A Monte Carlo study is then carried out which statistically verify the proximity of these optimal solutions to the global optimum, for the MIP problem.

关键词： Neural networks Network topology backpropagation algorithms Mathematical model Genetic algorithms Circuit topology Parallel processing Linear programming Multi-layer neural network Broadcasting

来源：评论

学校读者我要写书评

暂无评论

Generalization of the cross-entropy error function to improve the error backpropagation algorithm

Generalization of the cross-entropy error function to improv...

引用

International Conference on Neural Networks

作者： Sang-Hoon Oh Mobile Protocol and Signaling Section Mobile Communication System Division Electronics and Telecommunications Research Institute Taejon South Korea

ISBN: (纸本)0780341228

This paper generalizes the cross-entropy error function to improve the EBP (error back propagation) algorithm of multilayer perceptrons. The generalized error function reduces the probability that output nodes are near the wrong extreme value as well as the correct extreme value of sigmoid function. As a result, we can accelerate the learning speed of the EBP algorithm with improved generalization performance. The effectiveness of the proposed method is demonstrated in a handwritten digit recognition task.

关键词： backpropagation algorithms Multilayer perceptrons Acceleration Handwriting recognition Pattern recognition Error correction Iterative algorithms Protocols Mobile communication Databases

来源：评论

学校读者我要写书评

暂无评论

A matched-field backpropagation algorithm for source localization

A matched-field backpropagation algorithm for source localiz...

引用

OCEANS

作者： D.J. Thomson G.R. Ebbeson B.H. Maranda Esquimalt Defence Research DetachmentCFB Esquimalt Defence Research Establishment Victoria BC Canada Defence Research Establishment Dartmouth NS Canada

Model-based signal processing techniques have been developed over the years to improve the capability of active and passive sonar systems for detecting and localizing quiet underwater targets. In a generic matched-field processor, hydrophone signals measured at the array are compared to hypothetical signals (replicas) that are calculated by a full-field acoustic model for a given target position. This matching is carried out for many potential target locations within a search region (range, depth and bearing) to form an ambiguity surface whose peak values provide the greatest likelihood that targets are present. In this paper, we evaluate a version of a matched-field processor that combines measured data with a higher-order parabolic equation (PE) algorithm to effectively backpropagate an (unnormalized) ambiguity surface outwards from the receiving array. To illustrate this PE-based method, the unconventional processor is applied to some synthetic and experimental hydrophone data received on vertical line arrays in shallow-water waveguides.

关键词： backpropagation algorithms Signal processing algorithms Sonar detection Sonar equipment Signal processing Acoustic signal processing Array signal processing Sonar measurements Acoustic signal detection Underwater tracking

来源：评论

学校读者我要写书评

暂无评论

Synthetical backpropagation algorithm

Synthetical backpropagation algorithm

引用

IEEE International Conference on Intelligent Processing Systems, ICIPS

作者： Kejun Wang Hongzhang Jin Guobin Li Department of Automatic Control Harbin Engineering of Technology Harbin China

ISBN: (纸本)0780342534

The paper presents a synthetical backpropagation algorithm (SBP) for multilayer feedforward neural networks. The general ability, speed of training, and global optimal solution are considered in SBP. A new generalized error index function is used in SBP. The backpropagation technique based on searching output space is proposed and used in SBP. SBP has a dynamical adaptive regulation learning rate, a variable momentum coefficient and ability of self regulation activate characteristic. The contrast experiments indicate that SBP has a faster convergence speed than BP and can achieve a global optimal solution.

关键词： backpropagation algorithms Tin Automatic control Neural networks Feedforward neural networks Extraterrestrial measurements Algorithm design and analysis

来源：评论

学校读者我要写书评

暂无评论

A hybrid global learning algorithm based on global search and least squares techniques for backpropagation networks

A hybrid global learning algorithm based on global search an...

引用

International Conference on Neural Networks

作者： Chi-Tat Leung T.W.S. Chow Department of Electronic Engineering City University of Hong Kong Kowloon Hong Kong China

A hybrid learning algorithm for backpropagation network based on global search and least squares methods is presented to increase the speed of convergence. The proposed algorithm comprises global search and least squares parts. The global search part trains a backpropagation network over a reduced weight space. The remained weights are calculated in accordance with linear least squares method. Two problems of nonlinear function approximation and modified XOR are applied to demonstrate the fast global search performance of the proposed algorithm. The results indicate that the proposed algorithm enables the learning process to significantly speed up by at most 4670% in terms of iterations and do not trap in local minima.

关键词： Least squares methods Neurons backpropagation algorithms Search methods Convergence Supervised learning Function approximation Approximation algorithms Robots Nonlinear systems

来源：评论

学校读者我要写书评

暂无评论

/spl delta/-NARMA neural networks: a new approach to signal prediction

引用

IEEE Transactions on Signal Processing 1997年第11期45卷 2799-2810页

作者： D. Bonnet V. Labouisse A. Grumbach Prospective Research Department SNCF Paris France Paris France

This paper presents a new connectionist architecture for stochastic univariate signal prediction. After a review of related statistical and connectionist models pointing out their advantages and limitations, we introduce the /spl epsiv/-NARMA model as the simplest nonlinear extension of ARMA models. These models then provide the units of a MLP-like neural network: the /spl delta/-NARMA neural network. The associated learning algorithm is based on an extension of classical backpropagation and on the concept of virtual error. Such networks can be seen as an extension of ARIMA and ARARMA models and face the problem of nonstationary signal prediction. A theoretical study brings understanding of experimental phenomena observed during the /spl delta/-NARMA learning process. The experiments carried out on three railroad-related real-life signals suggest that /spl delta/-NARMA networks outperform other studied univariate models.

关键词： Neural networks Autoregressive processes Predictive models Signal processing Stochastic processes Neurofeedback Cognitive science Artificial intelligence backpropagation algorithms Multi-layer neural network

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：