ANALOG NOISE-ENHANCED LEARNING IN NEURAL NETWORK CIRCUITS

被引：18

作者：

MURRAY, AF

机构：

[1] Department of Electrical Engineering, University of Edinburgh, Edinburgh

来源：

ELECTRONICS LETTERS | 1991年 / 27卷 / 17期

关键词：

NETWORKS AND NETWORK THEORY;

D O I：

10.1049/el:19910970

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Experiments are reported which demonstrate that, whereas digital inaccuracy in neural arithmetic, in the form of bit-length limitation, degrades neural learning, analogue noise enhances it dramatically. The classification task chosen is that of vowel recognition within a multilayer perceptron network, but the findings seem to be perfectly general in the neural context, and have ramifications for all learning processes where weights evolve incrementally, and slowly.

引用

页码：1546 / 1548

页数：3