An efficient implementation of a learning method for Mamdani fuzzy models

Author

Schnitman, Leizer ; Yoneyama, Takashi

Author_Institution

Inst. Tecnologico de Aeronautica, Sao Jose dos Campos, Brazil

fYear

2000

fDate

2000

Firstpage

38

Lastpage

43

Abstract

This paper presents an efficient implementation of a supervised learning method based on membership function training in the context of Mamdani fuzzy models. The main idea is to adjust the antecedent and consequent membership functions that are of asymmetric trapezoidal form by backpropagating the output error through the fuzzy net. The proposed implementation is analogous to the training scheme commonly used with Takagi-Sugeno fuzzy models but it requires additional procedures that are related to some specific characteristics of the Mamdani fuzzy structures. Some numerical results are provided as illustrations

Keywords

backpropagation; computational complexity; fuzzy neural nets; learning (artificial intelligence); Mamdani fuzzy models; Takagi-Sugeno fuzzy models; antecedent membership function adjustment; asymmetric trapezoidal functions; consequent membership function adjustment; efficient implementation; fuzzy neural net; membership function training; output error backpropagation; supervised learning method; Backpropagation algorithms; Context modeling; Control system synthesis; Fuzzy neural networks; Learning systems; Mathematical model; Optimization methods; Power system modeling; Supervised learning; Takagi-Sugeno model;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 2000. Proceedings. Sixth Brazilian Symposium on

Conference_Location

Rio de Janeiro, RJ

ISSN

1522-4899

Print_ISBN

0-7695-0856-1

Type

conf

DOI

10.1109/SBRN.2000.889710

Filename

889710