检索结果-内蒙古大学图书馆

The dependence identification neural network construction algorithm

IEEE TRANSACTIONS ON NEURAL NETWORKS 1996年第1期7卷 3-15页

作者： Moody, JO Antsaklis, PJ Department of Electrical Engineering University of Notre Dame Notre Dame IN USA

An algorithm for constructing and training multilayer neural networks, dependence identification, is presented in this paper. Its distinctive features are that i) it transforms the training problem into a set of quadratic optimization problems that are solved by a number of linear equations, ii) it constructs an appropriate network to meet the training specifications, and iii) the resulting network architecture and weights can be further refined with standard training algorithms, like backpropagation, giving a significant speedup in the development time of the neural network and decreasing the amount of trial and error usually associated with network development.

关键词： Neural networks Neurons backpropagation algorithms Iterative algorithms Transforms Equations Standards development Multi-layer neural network Intelligent networks

来源：评论

学校读者我要写书评

暂无评论

Fuzzy parameter adaptation in optimization: Some neural net training examples

引用

IEEE COMPUTATIONAL SCIENCE & ENGINEERING 1996年第1期3卷 57-65页

作者： Arabshahi, P Choi, JJ Marks, RJ Caudell, TP UNIV WASHINGTON SEATTLEWA 98195 BOEING CO COMP SERVSEATTLEWA 98124 UNIV NEW MEXICO ALBUQUERQUENM 87131

Many types of artificial neural networks require training to "learn" the tasks they will be called upon to do, which basically translates into properly setting the weights of the interneuron connections or "synapses" so that the network will give the desired output. Values of certain parameters in the algorithms used to train the networks must be chosen and then adapted or optimized carefully. Error backpropagation training of the multilayer perceptron, for example, requires judicious choice of the step and momentum parameters. The ART network requires choice of a vigilance parameter, while a learning rate must be selected for Kohonen networks. These parameters are typically chosen and adapted by a human "neural smith" using heuristic rules. For example, a smooth error surface in the backpropagation training of a layered perceptron suggests use of a long step, whereas a steep surface indicates a need for smaller steps. The heuristic description of the parameter selection is fuzzy. The terms "smooth," "long," "steep," and "smaller" are each fuzzy linguistic variables that can be quantified into a fuzzy inference engine. A properly designed fuzzy controller, then, can relieve the neural smith of some labor, and get the job done quicker and better. The concept has broader possible applications, beyond neural network training, to parameter selection and optimization generally.

关键词： Fuzzy neural networks Neural networks Fuzzy control Fuzzy sets Fuzzy systems Artificial neural networks backpropagation algorithms Subspace constraints Humans Fuzzy logic

来源：评论

学校读者我要写书评

暂无评论

The cascade-correlation learning: A projection pursuit learning perspective

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS 1996年第2期7卷 278-289页

作者： Hwang, JN You, SS Lay, SR Jou, IC MINIST TRANSPORTAT & COMMUN TELECOMMUN LABS CHUNGLI 320 TAIWAN

Cascade-correlation (Cascor) is a popular supervised learning architecture that dynamically grows layers of hidden neurons of fixed nonlinear activations (e.g., sigmoids), so that the network topology (size, depth) can be efficiently determined. Similar to a cascade-correlation learning network (CCLN), a projection pursuit learning network (PPLN) also dynamically grows the hidden neurons. Unlike a CCLN where cascaded connections from the existing hidden units to the new candidate hidden unit are required to establish high-order nonlinearity in approximating the residual error, a PPLN approximates the high-order nonlinearity by using trainable parametric or semiparametric nonlinear smooth activations based on minimum mean squared error criterion. An analysis is provided to show that the maximum correlation training criterion used in a CCLN tends to produce hidden units that saturate and thus makes it more suitable for classification tasks instead of regression tasks as evidenced in the simulation results. It is also observed that this critical weakness in CCLN can also potentially carry over to classification tasks, such as the two-spiral benchmark used in the original CCLN paper.

关键词： Neurons Network topology Laboratories backpropagation algorithms Supervised learning Analytical models Multilayer perceptrons Biological neural networks Neural networks Multi-layer neural network

来源：评论

学校读者我要写书评

暂无评论

Optimization of HV electrode systems by neural networks using a new learning method

引用

IEEE TRANSACTIONS ON DIELECTRICS AND ELECTRICAL INSULATION 1996年第6期3卷 737-742页

作者： Mukherjee, PK Trinitis, C Steinbigler, H TECH UNIV MUNICH HIGH VOLTAGE INSTD-8000 MUNICHGERMANY

To avoid a large number of iterations, optimization of electrode shapes has been done by artificial neural networks (NN). Two practical examples have been considered, an axisymmetric single-phase GIS bus termination and an axisymmetric transformer shield ring. The shape of the electrodes has been taken as quarter-ellipse or half-ellipse because an ellipse has more flexibility a than circle. For NN, the socalled resilient propagation algorithm, learning faster than the standard back-propagation algorithm, has been employed. The training sets as well as the test sets of NN have been prepared by charge simulation method.

关键词： Optimization methods Electrodes Neural networks Learning systems Neurons Artificial neural networks Shape Computational modeling backpropagation algorithms Power transformer insulation

来源：评论

学校读者我要写书评

暂无评论

An optimal tracking neuro-controller for nonlinear dynamic systems

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS 1996年第5期7卷 1099-1110页

作者： Park, YM Choi, MS Lee, KY PENN STATE UNIV DEPT ELECT ENGNUNIVERSITY PKPA 16802

Multilayer neural networks are used to design an optimal tracking neuro-controller (OTNC) for discrete-time nonlinear dynamic systems with quadratic cost function, The OTNC is made of two controllers: feedforward neuro-controller (FFNC) and feedback neuro-controller (FBNC), The FFNC controls the steady-state output of the plant, while the FBNC controls the transient-state output of the plant, The FFNC is designed using a novel inverse mapping concept by using a neuro-identifier, A generalized backpropagation-through-time (GBTT) algorithm is developed to minimize the general quadratic cost function for the FBNC training, The proposed methodology is useful as an off-line control method where the plant is first identified and then a controller is designed for it, A case study for a typical plant with nonlinear dynamics shows good performance of the proposed OTNC.

关键词： Neural networks Cost function Optimal control Error correction backpropagation algorithms Neurofeedback Steady-state Power engineering and energy Distributed control Power generation

来源：评论

学校读者我要写书评

暂无评论

Dynamic recurrent neural networks: A dynamical analysis

引用

IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS 1996年第5期26卷 692-706页

作者： Draye, JPS Pavisic, DA Cheron, GA Libert, GA UNIV MONS DEPT NEUROSCI B-7000 MONS BELGIUM FREE UNIV BRUSSELS BIOMECH LAB B-1050 BRUSSELS BELGIUM

In this paper, we explore the dynamical features of a neural network model which presents two types of adaptative parameters: the classical weights between the units and the time constants associated with each artificial neuron, The purpose of this study is to provide a strong theoretical basis for modeling and simulating dynamic recurrent neural networks. In order to achieve this, we study the effect of the statistical distribution of the weights and of the time constants on the network dynamics and we make a statistical analysis of the neural transformation, We examine the network power spectra (to draw some conclusions over the frequential behavior of the network) and me compute the stability regions to explore the stability of the model, We show that the network is sensitive to the variations of the mean values of the weights and the time constants (because of the temporal aspects of the learned tasks), Nevertheless, our results highlight the improvements in the network dynamics due to the introduction of adaptative time constants and indicate that dynamic recurrent neural networks can bring new powerful features in the field of neural computing.

关键词： Recurrent neural networks Information processing Laboratories Artificial neural networks Biological system modeling Associative memory backpropagation algorithms Biology computing Multi-layer neural network Neural networks

来源：评论

学校读者我要写书评

暂无评论

Predicting sun spots using a layered perceptron neural network

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS 1996年第2期7卷 501-505页

作者： Park, YR Murray, TJ Chen, C TEXAS A&M UNIV COLL BUSINESS ADM LAREDO TX 78040 USA SYRACUSE UNIV SCH MANAGEMENT BRETHEN OPERAT MANAGEMENT INST SYRACUSE NY 13244 USA

Interest in neural networks has expanded rapidly in recent years, Selecting the best structure for a given task, however, remains a critical issue in neural-network design. Although the performance of a network clearly depends on its structure, the procedure for selecting the optimal structure has not been thoroughly investigated. It Is well known that the number of hidden units must be sufficient to discriminate each observation correctly. A large number of hidden units requires extensive computational time for training and often times prediction results may not be as accurate as expected. This study attempts to apply the principal component analysis (PCA) to determine the structure of a multilayered neural network for time series forecasting problems. The main focus is to determine the number of hidden units for a multilayered feedforward network. One empirical experiment with sunspot data is used to demonstrate the usefulness of the proposed approach.

关键词： Sun Neural networks Decision trees Statistics Input variables Training data backpropagation algorithms Distribution functions

来源：评论

学校读者我要写书评

暂无评论

A rapid supervised learning neural network for function interpolation and approximation

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS 1996年第5期7卷 1220-1230页

作者： Chen, CLP Department of Computer Science and Engineering Wright State University Dayton OH USA

This paper presents a neural-network architecture and an instant learning algorithm that rapidly decides the weights of the designed single-hidden layer neural network For an n-dimensional N-pattern training set, with a constant bias, a maximum of N - r - 1 hidden nodes is required to learn the mapping within a given precision (where r is the rank, usually the dimension, of the input patterns), For off-line training, the proposed network and algorithm is able to achieve ''one-shot'' training as opposed to most iterative training algorithms in the literature. An on-line training algorithm is also presented. Similar to most of the backpropagation type of learning algorithms, the given algorithm also interpolates the training data, To eliminate outlier data which mag appear in some erroneous training data, a robust weighted least squares method is proposed. The robust weighted least squares learning algorithm can eliminate outlier samples and the algorithm approximates the training data rather than interpolates them. The advantage of the designed network architecture is also mathematically proved. Several experiments show very promising results.

关键词： Supervised learning Neural networks Interpolation backpropagation algorithms Multi-layer neural network Computer architecture Least squares methods Computer science Computer networks

来源：评论

学校读者我要写书评

暂无评论

Multilayer neural-net robot controller with guaranteed tracking performance

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS 1996年第2期7卷 388-399页

作者： Lewis, FL Yesildirek, A Liu, K Automation and Robotics Research Institute University of Texas Arlington TX USA University of Texas at Arlington Arlington TX US Autom. & Robotics Res. Inst. Texas Univ. Arlington TX USA

A multilayer neural-net (NN) controller for a general serial-link rigid robot arm is developed. The structure of the NN controller is derived using a filtered error/passivity approach. No off-line learning phase is needed for the proposed NN controller and the weights are easily initialized. The nonlinear nature of the NN, plus NN functional reconstruction inaccuracies and robot disturbances, mean that the standard delta rule using backpropagation tuning does not suffice for closed-loop dynamic control. Novel on-line weight tuning algorithms, including correction terms to the delta rule plus an added robustifying signal, guarantee bounded tracking errors as well as bounded NN weights. Specific bounds are determined, and the tracking error bound can be made arbitrarily small by increasing a certain feedback gain. The correction terms involve a second-order forward-propagated wave in the backprop network. New NN properties including the notions of a passive NN, a dissipative NN, and a robust NN are introduced.

关键词： Nonhomogeneous media Robot control Neural networks Neurons Robust control backpropagation algorithms Performance analysis Robust stability Robot sensing systems Nonlinear dynamical systems

来源：评论

学校读者我要写书评

暂无评论

Accelerating the training of feedforward neural networks using generalized Hebbian rules for initializing the internal representations

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS 1996年第2期7卷 419-426页

作者： Karayiannis, NB Department of Electrical and Computer Engineering University of Houston TX USA

This paper presents an unsupervised learning scheme for initializing the internal representations of feedforward neural networks, which accelerates the convergence of supervised learning algorithms. It is proposed in this paper that the initial set of internal representations can be formed through a bottom-up unsupervised learning process applied before the top-down supervised training algorithm. The synaptic weights that connect the input of the network with the hidden units can be determined through linear or nonlinear variations of a generalized Hebbian learning rule, known as Oja's rule. Various generalized Hebbian rules were experimentally tested and evaluated in terms of their effect on the convergence of the supervised training process. Several experiments indicated that the use of the proposed initialization of the internal representations significantly improves the convergence of gradient-descent-based algorithms used to perform nontrivial training tasks. The improvement of the convergence becomes significant as the size and complexity of the training task increase.

关键词： Acceleration Neural networks Feedforward neural networks backpropagation algorithms Convergence Application software Unsupervised learning Multi-layer neural network Entropy Supervised learning

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：