检索结果-内蒙古大学图书馆

Seminar on Neural Network Applications in Electrical Engineering

作者： D.P. Mandic I.R. Krcmar School of Information Systems University of East Anglia Norwich UK Faculty of Electrical Engineering University of Banja Luka Banja Luka Bosnia and Herzegovina

ISBN: (纸本)0780355121

Relationships between the learning rate /spl eta/ and the slopes /spl beta/ in the tanh activation function for a feedforward neural network (NN) are provided. The analysis establishes the equivalence in the static and dynamic sense between a referent and an arbitrary feedforward NN which helps to reduce the number of degrees of freedom in learning algorithms for NNs.

关键词： Neurons Neural networks Signal processing algorithms backpropagation algorithms Feedforward neural networks Error correction Algorithm design and analysis Computational complexity System identification Process control

来源：评论

学校读者我要写书评

暂无评论

A neural network sidelobe suppression filter for a pulse-compression radar with powers-of-two weights

A neural network sidelobe suppression filter for a pulse-com...

引用

Mediterranean Electrotechnical Conference (MELECON)

作者： Yu.P. Grishin A. Zankiewicz Department of Electrical Engineering Bialystok Technical University Poland

A new compression filter for binary coded wideband digital signals with sidelobe reduction capabilities using a feedforward time-delay neural network is considered. This filter uses powers-of-two synaptic weights and the backpropagation learning algorithm. Using digital hardware implementation of such a filter for pulse compression is easier owing to elimination of multipliers. The simulation results showed that the time sidelobes of the output signal can be reduced up to about 60 dB. The quantization of the filter weights causes additional losses of about 6-14 dB depending on the input signal to noise ratio.

关键词： Neural networks Radar Digital filters Wideband Feedforward neural networks backpropagation algorithms Hardware Pulse compression methods Quantization Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Neural network face recognition using statistical feature extraction

Neural network face recognition using statistical feature ex...

引用

National Conference on Radio Science (NSRC)

作者： S.E. El-Khamy O.A. Abdel-Alim M.M. Saii Department of Electrical Engineering Faculty of Engineering Alexandria University Alexandria Egypt

Recognition method of human face using statistical analysis feature extraction and a neural network algorithm is proposed. In the preprocessing step we detect the edges of the face image by using the Sobel algorithm. Then we propose a new method to transform the two-dimension black and white image to a one-dimension vector. Finally, based on the statistical analysis, we extract seven features. In the recognition step we use the fast backpropagation (FBP) algorithm. Computer simulation results with 100 test images of 10 persons (the images of each person in a various pauses, facial expression, and facial details) show that the proposed method yields a high recognition rate.

关键词： Neural networks Face recognition Face detection Statistical analysis Feature extraction Image edge detection backpropagation algorithms Humans Computer simulation Testing

来源：评论

学校读者我要写书评

暂无评论

A chemical reactor benchmark for parallel adaptive control using feedforward neural networks

A chemical reactor benchmark for parallel adaptive control u...

引用

Brazilian Symposium on Neural Networks

作者： C.O. Cajueiro E.M. Hemerly ITA-IEE-IEES Instituto Tecnológico de Aeronautica Sao Paulo Brazil

This paper applies a parallel scheme for adaptive control that uses only one neural network to a CSTR (continuous stirred tank reactor). Convergence of the identification error is investigated by Lyapunov's second method. The training process of the neural network is carried out by using two different techniques: backpropagation and extended Kalman filter algorithm.

关键词： Chemical reactors Adaptive control Neural networks Feedforward neural networks Uncertainty Continuous-stirred tank reactor Testing backpropagation algorithms Network topology Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Neural metrics-software metrics in artificial neural networks

Neural metrics-software metrics in artificial neural network...

引用

International Conference on Knowledge-Based Intelligent Information Engineering Systems (KES)

作者： W.K. Leung R. Simpson School of Computing Faculty of Computing Information and English University of Central England UK

backpropagation based supervised feedforward artificial neural networks (ANNs) have been developed for many applications (e.g. Rumelhart et al., 1986; Hinton, 1989; Werbos, 1990; and Riedmiller, 1994) but no detailed study of the measurement of the quality characteristics (e.g complexity and efficiency) of the network system has been made. Without an appropriate measurement, it is difficult to tell how the network performs on given applications. In addition, it is difficult to provide a measure of the algorithmic complexity of any given application. The paper proposes a new set of software metrics, named neural metrics, which provide indicative measures of the quality characteristics of ANNs. Neural metrics that are non-primitive in nature are defined mathematically as neural metrics functions.

关键词： Software metrics Intelligent networks Artificial neural networks backpropagation algorithms Neural networks Application software Training data Computer networks Logic testing Performance evaluation

来源：评论

学校读者我要写书评

暂无评论

Support vector machine for regression and applications to financial forecasting

Support vector machine for regression and applications to fi...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： T.B. Trafalis H. Ince School of Industrial Engineering University of Oklahama Norman OK USA

The main purpose of the paper is to compare the support vector machine (SVM) developed by Cortes and Vapnik (1995) with other techniques such as backpropagation and radial basis function (RBF) networks for financial forecasting applications. The theory of the SVM algorithm is based on statistical learning theory. Training of SVMs leads to a quadratic programming (QP) problem. Preliminary computational results for stock price prediction are also presented.

关键词： Support vector machines Virtual colonoscopy Statistical learning Quadratic programming Support vector machine classification Industrial engineering backpropagation algorithms Machine learning Machine learning algorithms Pattern classification

来源：评论

学校读者我要写书评

暂无评论

Preprocessing for handwritten Chinese character based on neural networks

Preprocessing for handwritten Chinese character based on neu...

引用

World Congress on Intelligent Control and Automation (WCICA)

作者： Yan Peng Yan Zuquan Second Artillery Institute of Engineering Xi'an China

Introduces the entropy concept to a backpropagation neural network in order to improve its performance of classifying similar samples. Then, based on the neural network, we propose a thinning model for Chinese characters, and state its preprocessing methods, neural network architecture, learning algorithm and working algorithm. At the same time, we reform the conventional thinning process for Chinese characters in order to improve speed of processing. Finally, combining the nonlinear shape normalization methods and the reformed thinning process, an experiment on handwritten Chinese character preprocessing has been done to test the model, and obtain satisfactory results.

关键词： Neural networks Entropy backpropagation algorithms Shape Testing

来源：评论

学校读者我要写书评

暂无评论

Forecasting time series using logical combinations of neural-based networks

Forecasting time series using logical combinations of neural...

引用

IEEE International Conference on Systems, Man and Cybernetics

作者： A. Thammano Advanced Computer Application and Design Research Group Faculty of Information Technology King Mongkut's Institute of Technology Ladkrabang Bangkok Thailand

Forecasts are the basis for planning and decision making. The more accurate the organization's forecasts, the better prepared it will be to take advantage of future opportunities and to reduce potential risks. Thus, it should come as no surprise that there is a tremendous number of statistical prediction algorithms already in existence. However, most of them are useful only when the time series exhibit little trend or seasonal variations but a great deal of irregular or random variation. Therefore, the objective of this paper is to propose a new intelligent forecasting technique which is constructed by combining n-trained neural-based network together. The experimental results based on simulated data show a significant improvement in forecasting accuracy.

关键词： Neural networks Predictive models Fuzzy systems backpropagation algorithms Economic forecasting Feedforward neural networks Intelligent networks Decision making Econometrics Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

backpropagation in turbid media

引用

JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION 1999年第6期16卷 1254-1265页

作者： Matson, CL Liu, HL USAF Res Lab Adv Opt & Imaging Div Kirtland AFB NM 87117 USA Univ Texas Dept Biomed Engn Arlington TX 76019 USA

We extend the backpropagation algorithm of standard diffraction tomography to backpropagation in turbid media. We analyze the behavior of the backpropagation algorithm both for a single-view geometry, as is common in mammography, and for multiple views. The most general form of the algorithm permits arbitrary placement of sources and detectors in the background medium. In addition, we specialize the algorithm for the case of a planar array of detectors, which permits the backpropagation algorithm to be implemented with fast-Fourier-domain noniterative algebraic methods. In this case the algorithm can be used to reconstruct three-dimensional images in a minute or less, depending on the number of views. We demonstrate the theoretical results with computer simulations. (C) 1999 Optical Society of America [S0740-3232(99)02806-9].

关键词： Three-Dimensional Images backpropagation algorithms algorithms Minute back propagation View Planar antennas

来源：评论

学校读者我要写书评

暂无评论

Improving the convergence of the backpropagation algorithm using learning rate adaptation methods

引用

NEURAL COMPUTATION 1999年第7期11卷 1769-1796页

作者： Magoulas, GD Vrahatis, MN Androulakis, GS Univ Athens Dept Informat GR-15771 Athens Greece Univ Patras Artificial Intelligence Res Ctr GR-26110 Patras Greece Univ Patras Dept Math GR-26110 Patras Greece

This article focuses on gradient-based backpropagation algorithms that use either a common adaptive learning rate for all weights or an individual adaptive learning rate for each weight and apply the Goldstein/Armijo line search. The learning-rate adaptation is based on descent techniques and estimates of the local Lipschitz constant that are obtained without additional error function and gradient evaluations. The proposed algorithms improve the backpropagation training in terms of both convergence rate and convergence characteristics, such as stable learning and robustness to oscillations. Simulations are conducted to compare and evaluate the convergence behavior of these gradient-based training algorithms with several popular training methods.

关键词： backpropagation algorithms additional error learning rate Converge

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：