Rates of convergence of the recursive radial basis function networks

Author

Mazurek, J. ; Krzyzak, A. ; Cichocki, Andrzej

Author_Institution

Neurolab. GmbH, Germany

Volume

4

fYear

1997

fDate

21-24 Apr 1997

Firstpage

3317

Abstract

Recursive radial basis function (RRBF) neural networks are introduced and discussed. We study in detail the nets with diagonal receptive field matrices. Parameters of the networks are learned by a simple procedure. Convergence and the rates of convergence of RRBF nets in the mean integrated absolute error (MIAE) sense are studied under mild conditions imposed on some of the network parameters. The obtained results also give the upper bounds on the performance of RRBF nets learned by minimizing the empirical L₁ error

Keywords

adaptive systems; approximation theory; convergence of numerical methods; error analysis; feedforward neural nets; learning (artificial intelligence); matrix algebra; network parameters; recursive functions; signal processing; adaptive learning algorithms; convergence rates; diagonal receptive field matrices; empirical L₁ error minimization; function approximation; mean integrated absolute error; network parameters; neural networks; performance; processing nodes; recursive radial basis function networks; signal processing; upper bounds; Convergence; Radial basis function networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1997. ICASSP-97., 1997 IEEE International Conference on

Conference_Location

Munich

ISSN

1520-6149

Print_ISBN

0-8186-7919-0

Type

conf

DOI

10.1109/ICASSP.1997.595503

Filename

595503