NEW ALGORITHMS FOR TRAINING FEEDFORWARD NEURAL NETWORKS

被引：20

作者：

KAK, S

机构：

[1] Department of Electrical and Computer Engineering, Louisiana State University, Baton Rouge

来源：

PATTERN RECOGNITION LETTERS | 1994年 / 15卷 / 03期

关键词：

FEEDFORWARD NEURAL NETWORKS; CORNER CLASSIFICATION; GENERALIZATION; MACHINE LEARNING;

D O I：

10.1016/0167-8655(94)90062-0

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A new network that maps n-dimensional binary vectors into m-dimensional binary vectors using 3-layered feedforward neural networks is described. Algorithms to train this network are presented. The computing power of the new algorithms may be gauged from the example that the exclusive-Or problem that requires several thousand iterative steps using the backpropagation algorithm was solved in eight or fewer steps.

引用

页码：295 / 298

页数：4