NUMERICAL RELATIONSHIPS BETWEEN NEURAL NETWORKS, CONTINUOUS-FUNCTIONS, AND FUZZY-SYSTEMS

被引：25

作者：

BUCKLEY, JJ

HAYASHI, Y

机构：

[1] Mathematics Department, University of Alabama at Birmingham, Birmingham

[2] Department of Computer and Information Sciences, Ibaraki University, Hitachi, Ibaraki

来源：

FUZZY SETS AND SYSTEMS | 1993年 / 60卷 / 01期

关键词：

NEURAL NETWORKS; FUZZY SYSTEMS; FUZZY CONTROLLER; FUZZY EXPERT SYSTEMS;

D O I：

10.1016/0165-0114(93)90283-N

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

In this survey paper we first discuss computational equivalence between continuous functions, regular neural nets, fuzzy controllers, and discrete fuzzy expert systems. We also discuss how to build hybrid neural nets numerically identical to a fuzzy controller or a discrete fuzzy expert system. We then discuss computational equivalence between continuous fuzzy functions, regular fuzzy neural nets, and fuzzy systems. We also discuss how to construct hybrid fuzzy neural nets to approximate continuous fuzzy functions.

引用

页码：1 / 8

页数：8

共 57 条

[41] APPROXIMATION OF CONTINUOUS-FUNCTIONS ON R(D) BY LINEAR-COMBINATIONS OF SHIFTED ROTATIONS OF A SIGMOID FUNCTION WITH AND WITHOUT SCALING [J].