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

Authors
BUCKLEY JJ HAYASHI Y
Citation
Jj. Buckley et Y. Hayashi, NUMERICAL RELATIONSHIPS BETWEEN NEURAL NETWORKS, CONTINUOUS-FUNCTIONS, AND FUZZY-SYSTEMS, Fuzzy sets and systems, 60(1), 1993, pp. 1-8
Citations number
58
Categorie Soggetti
Computer Sciences, Special Topics","System Science",Mathematics,"Computer Applications & Cybernetics","Statistic & Probability",Mathematics
Journal title
Fuzzy sets and systems → ACNP
ISSN journal
01650114
Volume
60
Issue
1
Year of publication
1993
Pages
1 - 8
Database
ISI
SICI code
0165-0114(1993)60:1<1:NRBNNC>2.0.ZU;2-K
Abstract
In this survey paper we first discuss computational equivalence betwee n continuous functions, regular neural nets, fuzzy controllers, and di screte fuzzy expert systems. We also discuss how to build hybrid neura l nets numerically identical to a fuzzy controller or a discrete fuzzy expert system. We then discuss computational equivalence between cont inuous fuzzy functions, regular fuzzy neural nets, and fuzzy systems. We also discuss how to construct hybrid fuzzy neural nets to approxima te continuous fuzzy functions.