DocumentCode
3224125
Title
Neural nets can be universal approximators for fuzzy functions
Author
Buckley, J.J. ; Hayashi, Yoichi
Author_Institution
Dept. of Math., Alabama Univ., Birmingham, AL, USA
Volume
4
fYear
1997
fDate
9-12 Jun 1997
Firstpage
2347
Abstract
We first argue that the extension principle is too computationlly involved to be an efficient way for a computer to evaluate fuzzy functions. We then suggest using α-cuts and interval arithmetic to compute the values of fuzzy functions. Using this method of computing fuzzy functions, we then show that neural nets are universal approximators for (computable) fuzzy functions, when we only input non-negative, or non-positive, fuzzy numbers
Keywords
arithmetic; feedforward neural nets; function approximation; fuzzy set theory; multilayer perceptrons; α-cuts; fuzzy functions; interval arithmetic; nonnegative fuzzy numbers; nonpositive fuzzy numbers; universal approximator; Arithmetic; Computer science; Feedforward neural networks; Fuzzy neural networks; Fuzzy sets; Mathematics; Neural networks;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks,1997., International Conference on
Conference_Location
Houston, TX
Print_ISBN
0-7803-4122-8
Type
conf
DOI
10.1109/ICNN.1997.614430
Filename
614430
Link To Document