检索结果-内蒙古大学图书馆

fast learning algorithm for feed forward neural networks

Jisuanji Gongcheng/Computer Engineering 2005年第10期31卷 142-144+176页

作者： Jia, Wenchen Ye, Shiwei School of Distance and Continuing Education Grad. Sch. Chinese Academy of Sci. Beijing 100039 China School of Information Sci. and Engineering Grad. Sch. Chinese Academy of Sci. Beijing 100039 China

This paper presents a new algorithm for feed forward neural networks based on a new optimal target function constructed according to Young inequality in the conjugate of convex function. The optimal target function is a convex function to the weight values between neurons in different layers and outputs of hidden layers. First, outputs of hidden layers can be optimized with convex optimization method, then the weight values are optimized rapidly. The numerical experiments show that the new algorithm is very simple, can accelerate the convergence rate, strength the generalization capability and reduce the error.

关键词： Convex function fast learning algorithm Feed forward neural networks Young inequality

来源：评论

学校读者我要写书评

暂无评论

A fast learning algorithm with Promising Convergence Capability

A Fast Learning Algorithm with Promising Convergence Capabil...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Cheung, Chi-Chung Ng, Sin-Chun Lui, Andrew K. Xu, Sean Shensheng Hong Kong Polytech Univ Dept Elect & Informat Engn Hong Kong Hong Kong Peoples R China Open Univ Sch Sci Technol Hong Kong Peoples R China

ISBN: (纸本)9781424496365

Backpropagation (BP) learning algorithm is the most widely supervised learning technique which is extensively applied in the training of multi-layer feed-forward neural networks. Many modifications of BP have been proposed to speed up the learning of the original BP. However, these modifications sometimes cannot converge properly due to the local minimum problem. This paper proposes a new algorithm, which provides a systematic approach to make use of the characteristics of different fast learning algorithms so that the convergence of a learning process is promising with a fast learning rate. Our performance investigation shows that the proposed algorithm always converges with a fast learning rate in two popular complicated applications whereas other popular fast learning algorithms give very poor global convergence capabilities in these two applications.

关键词： Breast cancer Convergence Iris Iris recognition Neural networks Neurons Switches backpropagation backpropagation learning algorithm fast learning algorithm feedforward neural nets learning technique local minimum problem multilayer feed-forward neural networks promising convergence capability

来源：评论

学校读者我要写书评

暂无评论

fast learning algorithms for self-organizing map employing rough comparison WTA and its digital hardware implementation

引用

IEICE TRANSACTIONS ON ELECTRONICS 2004年第11期E87C卷 1787-1794页

作者： Tamukoh, H Horio, K Yamakawa, T Kyushu Inst Technol Grad Sch Life Sci & Syst Engn Kitakyushu Fukuoka 8080196 Japan

This paper describes a new fast learning algorithm for Self-Organizing Map employing a "rough comparison winner-take-all" and its digital hardware architecture. In rough comparison winner-take-all algorithm, the winner unit is roughly and strictly assigned in early and later learning stage, respectively. It realizes both of high accuracy and fast learning. The digital hardware of the self-organizing map with proposed WTA algorithm is implemented using FPGA. Experimental results show that the designed hardware is superior to other hardware with respect to calculation speed.

关键词： self-organizing map fast learning algorithm rough comparison winner-take-all hardware architecture

来源：评论

学校读者我要写书评

暂无评论

The Autoencoder Based on Generalized Neo-Fuzzy Neuron and its fast learning for Deep Neural Networks 2

The Autoencoder Based on Generalized Neo-Fuzzy Neuron and it...

引用

2nd IEEE International Conference on Data Stream Mining and Processing (DSMP)

作者： Bodyanskiy, Yevgeniy Peleshko, Dmytro Rashkevych, Yuriy Vynokurova, Olena Kharkiv Natl Univ Radio Elect Control Syst Res Lab Kharkov Ukraine IT Step Univ Lvov Ukraine Minist Educ & Sci Ukraine Kiev Ukraine Kharkiv Natl Univ Radio Elect Kharkov Ukraine

ISBN: (纸本)9781538628744

In this paper the autoencoder based on the generalized neo-fuzzy neurons is proposed. Also its fast learning algorithm based on quadratic criterion was proposed. Such system can be used as part of deep learning systems. The proposed autoencoder is characterized by high learning speed and less number of tuned parameters in comparison with well-known autoencoders of "bottle neck" type. The efficiency of proposed approach has been justified based on different benchmarks and real data sets.

关键词： autoencoder deep learning network neo-fuzzy neuron fast learning algorithm data compression

来源：评论

学校读者我要写书评

暂无评论

A fast learning Fully Complex-valued Relaxation Network (FCRN)

A Fast Learning Fully Complex-valued Relaxation Network (FCR...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Suresh, S. Savitha, R. Sundararajan, N. Nanyang Technol Univ Sch Comp Engn Singapore Singapore SJCE Mysore Karnataka India

ISBN: (纸本)9781424496365

This paper presents a fast learning algorithm for a single hidden layer complex-valued neural network named as the "Fully Complex-valued Relaxation Network (FCRN)". FCRN employs a fully complex-valued Gaussian like activation function (sech) in the hidden layer and an exponential activation function in the output layer. FCRN estimates the minimum energy state of a logarithmic error function which represents both the magnitude and phase errors explicitly to compute the optimum output weights for randomly chosen hidden layer parameters. As the weights are computed by the inversion of a nonsingular matrix, FCRN requires lesser computational effort during training. Performance studies using a synthetic function approximation problem and a QAM equalization problem show improved approximation ability of the proposed FCRN network.

关键词： Approximation algorithms Energy states Function approximation Neurons QAM equalization problem Quadrature amplitude modulation Training exponential activation function fast learning algorithm fully complex-valued Gaussian like activation function fully complex-valued relaxation network function approximation hidden layer complex-valued neural network learning (artificial intelligence) logarithmic error function matrix inversion neural nets nonsingular matrix inversion synthetic function approximation problem

来源：评论

学校读者我要写书评

暂无评论

A fast algorithm for AR parameter estimation using a novel noise-constrained least-squares method

引用

NEURAL NETWORKS 2010年第3期23卷 396-405页

作者： Xia, Youshen Kamel, Mohamed S. Leung, Henry Fuzhou Univ Coll Math & Comp Sci Fuzhou Peoples R China Univ Waterloo Dept Elect & Comp Engn Waterloo ON N2L 3G1 Canada Univ Calgary Dept Elect & Comp Engn Calgary AB T2N 1N4 Canada

In this paper, a novel noise-constrained least-squares (NCLS) method for online autoregressive (AR) parameter estimation is developed under blind Gaussian noise environments, and a discrete-time learning algorithm with a fixed step length is proposed. It is shown that the proposed learning algorithm converges globally to an AR optimal estimate. Compared with conventional second-order and high-order statistical algorithms, the proposed learning algorithm can obtain a robust estimate which has a smaller mean-square error than the conventional least-squares estimate. Compared with the learning algorithm based on the generalized least absolute deviation method, instead of minimizing a non-smooth linear L-1 function, the proposed learning algorithm minimizes a quadratic convex function and thus is suitable for online parameter estimation. Simulation results confirm that the proposed learning algorithm can obtain more accurate estimates with a fast convergence speed. (C) 2009 Elsevier Ltd. All Fights reserved.

关键词： Autoregressive parameter estimation Noise-constrained least-squares method fast learning algorithm Fixed step length Robust estimate Mean-square error

来源：评论

学校读者我要写书评

暂无评论

A Novel Style Takagi-Sugeno-Kang Fuzzy Classifier With Its fast Training on Style Data

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第4期34卷 2814-2831页

作者： Gu, Suhang Chung, Fu-Lai Wang, Shitong Changshu Inst Technol Sch Elect Engn & Automat Jiangsu 215500 Peoples R China Jiangnan Univ Sch AI & Comp Sci Wuxi 214122 Peoples R China Hong Kong Polytech Univ Dept Comp Hong Kong Peoples R China Wenzhou Kean Univ Dept Comp Sci Wenzhou 325015 Peoples R China

The classification of style data generally depends on both physical features of data and distinct styles originating from their own homogeneities. As a first attempt, a novel style Takagi-Sugeno-Kang (TSK) fuzzy classifier called STSK is developed in this study. STSK as an interpretable classifier consists of interpretable fuzzy rules with powerful representation ability on explicit physical features, implicit style features and the proposed style matrix of style data. By separating all features including the implicit style features in the antecedent parts of all fuzzy rules into fixed fuzzy partitions, followed by randomly assigned Gaussian membership functions with interpretable linguistic terms, the interpretable antecedents and hence the interpretable fuzzy rules can be assured for STSK. The training of STSK can be accomplished by solving the designed objective function through iteratively augmenting style features and updating the consequent parameters (including the style matrices) of all fuzzy rules in the form of their analytical solutions. Furthermore, by means of the Sherman-Morrison and Schur complement formulas, a fast learning algorithm F-STSK is derived to speed up the training of STSK to ensure at least comparable classification performance and simultaneously avoid successive yet time-consuming matrix inversion calculations caused by iterative style feature augmentation. Extensive experimental results on benchmark datasets validate the effectiveness of the proposed fuzzy classifier STSK and its fast learning algorithm. Moreover, six case studies about style data demonstrate the superiority of STSK over the comparison methods. The code of this study can be downloaded from https://***/gusuhang10/STSK.

关键词： TSK fuzzy classification style data style matrix style feature augmentation fast learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Airborne sonar target recognition using artificial neural network

引用

MATHEMATICAL AND COMPUTER MODELLING 2002年第3-4期35卷 429-440页

作者： Liang, M Palakal, MJ Indiana Univ Purdue Univ Dept Comp & Informat Sci Indianapolis IN 46202 USA Lucent Technol Inc Holmdel NJ 07733 USA

Airborne sonar target recognition involves two key technical issues: target feature extraction and classification. In this paper, the issue designing a feature classifier with high classification accuracy is discussed. Generally, the multilayered feed-forward neural network can be applied to the airborne sonar target feature classification to achieve the high performance requirement. However, neural networks trained by conventional error back-propagation (B-P) learning algorithms suffer from slow convergence rate and inadequate generalization ability, Detailed analysis of the B-P algorithm reveals that these problems are mainly related to the magnitudes of the components of the gradient vector and the direction of the vector associated with the severely ill-conditioned nature of the Hessian matrix of the error function. A fast back-propagation (F-BP) algorithm is therefore developed to accelerate the learning speed of the B-P algorithm. A dynamic training strategy is then applied to the F-BP algorithm to improve the generalization ability. Experiments are carried out for airborne sonar target feature classification using these algorithms. The results show that the performance of the neural network classifier trained with the proposed algorithm is superior to that of traditional B-P algorithm with a seven-fold learning speed advantage over B-P. (C) 2002 Elsevier Science Ltd. All rights reserved.

关键词： target recognition neural network learning algorithm dynamic training fast learning algorithm generalization nonparameter classifier

来源：评论

学校读者我要写书评

暂无评论

An adaptive noise reduction filter for discrete signal by use of sandglass-type neural network

引用

ELECTRICAL ENGINEERING IN JAPAN 1999年第4期127卷 39-51页

作者： Yoshimura, H Shimizu, T Isu, N Sugata, K Tottori Univ Tottori Japan

An adaptive noise reduction filter composed of a sandglass-type neural network (SNN) noise reduction filter (RF) is proposed in this paper. SNN was originally devised to work effectively for information compression. It is a hierarchial network and is symmetrically structured. SNN consists of the same number of units in the input and output layers and a smaller number of units in the hidden layer. It is known that SNN has signal processing performance which is equivalent to Karhunen-Loeve expansion after learning. We proved the theoretical suitability of SNN for an adaptive noise reduction filter for discrete signals. The SNNRF behaves optimally when the number of units in the hidden layer is equal to the rank of the covariance matrix of the signal components included in the input signal. Further we show by applying the recursive least squares method to learning of the SNNRF that the filter can process signals for on-line adaptive noise reduction. This is an extremely desirable feature for practical application. In order to verify the validity of SNNRF, we performed computer experiments examining how the noise reduction ability of SNNRF is affected by altering the properties of the input pattern, learning algorithm, and SNN. The results confirm that the SNNRF acquired appropriate characteristics for noise reduction from the input signals, and markedly improved the SNR of the signals. (C) 1999 Scripta Technica.

关键词： neural network Karhunen-Loeve expansion adaptive filter on-line processing noise reduction fast learning algorithm discrete signal

来源：评论

学校读者我要写书评

暂无评论

Construction of noise reduction filter by use of Sandglass-type Neural network

引用

IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES 1997年第8期E80A卷 1384-1390页

作者： Yoshimura, H Shimizu, T Isu, N Sugata, K Faculty of Engineering Tottori University Tottori-shi 680 Japan

A noise reduction filter composed of a sandglass-type neural network (Sandglass-type Neural network Noise Reduction Filter: SNNRF) was proposed in the present paper. Sandglass-type neural network (SNN) has symmetrical layer construction, and consists of the same number of units in input and output layers and less number of units in a hidden layer. It is known that SNN has the property of processing signals which is equivalent to KL expansion after learning. We applied the recursive least square (RLS) method to learning of SNNRF, so that the SNNRF became able to process on-line noise reduction. This paper showed theoretically that SNNRF behaves most optimally when the number of units in the hidden layer is equal to the rank of covariance matrix of signal component included in input signal. Computer experiments confirmed that SNNRF acquired appropriate char acteristics for noise reduction from input signals, and remarkably improved the SN ratio of the signals.

关键词： neural network K-L expansion on-line processing noise reduction fast learning algorithm

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：