A NEW FAST ALGORITHM FOR EFFECTIVE TRAINING OF NEURAL CLASSIFIERS

被引：3

作者：

CHOU, WS ^{[1
]}

CHEN, YC ^{[1
]}

机构：

[1] NATL TSING HUA UNIV,DEPT ELECT ENGN,HSINCHU 30043,TAIWAN

来源：

PATTERN RECOGNITION | 1992年 / 25卷 / 04期

关键词：

PATTERN CLASSIFICATION; NEURAL NETWORK; FAST ALGORITHM; EFFECTIVE LEARNING; METHOD OF POTENTIAL FUNCTIONS;

D O I：

10.1016/0031-3203(92)90090-6

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A neural classifier whose training can be executed very effectively is proposed to overcome the disadvantages of the method of potential functions. The power of the method of potential functions is limited by the severe requirements for computation time and storage. After mapping the computational structure of the method of potential functions to a three-layered feedforward network, a new fast learning algorithm is applied to train the net. By the characteristics of the new training algorithm, the network's complexity can be largely reduced without degrading its performance too much. That is to say, the algorithm will enable the network to learn in a very fast and, moreover, very effective manner.

引用

页码：423 / 429

页数：7