Levenberg-Marquardt method for ANFIS learning

Author

Jang, Jyh-Shing Roger ; Mizutani, Eiji

Author_Institution

Dept. of Comput. Sci., Tsinghua Univ., Hsinchu, China

fYear

1996

fDate

19-22 Jun 1996

Firstpage

87

Lastpage

91

Abstract

Presents the results of applying the Levenberg-Marquardt method (K. Levenberg, 1944, and D.W. Marquardt, 1963), which is a popular nonlinear least-squares method, to the ANFIS (Adaptive Neuro-Fuzzy Inference System) architecture proposed by Jang (IEEE Trans. on Systems, Man and Cybernctics, vol. 23, no. 3, pp 665-685, May 1993). Through empirical studies, we discuss the strengths and weaknesses of using such an efficient nonlinear regression technique for neuro-fuzzy modeling, and explain the tradeoffs between mapping precision and membership function interpretability

Keywords

adaptive systems; fuzzy neural nets; inference mechanisms; learning (artificial intelligence); least squares approximations; neural net architecture; ANFIS architecture; Adaptive Neuro-Fuzzy Inference System; Levenberg-Marquardt method; learning; mapping precision; membership function interpretability; neurofuzzy modelling; nonlinear least-squares method; nonlinear regression technique; Ear; Electrical capacitance tomography; Fuzzy neural networks; Fuzzy sets; Fuzzy systems; Learning systems; Least squares methods; Newton method; Recursive estimation; Tin;

fLanguage

English

Publisher

ieee

Conference_Titel

Fuzzy Information Processing Society, 1996. NAFIPS., 1996 Biennial Conference of the North American

Conference_Location

Berkeley, CA

Print_ISBN

0-7803-3225-3

Type

conf

DOI

10.1109/NAFIPS.1996.534709

Filename

534709