FEEDBACK NEURAL NETWORKS - NEW CHARACTERISTICS AND A GENERALIZATION

被引：9

作者：

KAK, SC

机构：

[1] Department of Electrical and Computer Engineering, Louisiana State University, Baton Rouge, 70803, Louisiana

来源：

CIRCUITS SYSTEMS AND SIGNAL PROCESSING | 1993年 / 12卷 / 02期

关键词：

D O I：

10.1007/BF01189877

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：

New characteristics of feedback neural networks are studied. We discuss in detail the question of updating of neurons given incomplete information about the state of the neural network. We show how the mechanism of self-indexing for such updating provides better results than assigning 'don't know' values to the missing parts of the state vector. Issues related to the choice of the neural model for a feedback network are also considered. Properties of a new complex valued neuron model that generalizes McCulloch-Pitts neurons are examined.

引用

页码：263 / 278

页数：16

共 18 条

[1] NEURAL THEORY OF ASSOCIATION AND CONCEPT-FORMATION [J].