Improved generalization learning with sliding mode control and the Levenberg-Marquadt algorithm

Author

Costa, Marcelo Azevedo ; Braga, Antônio Pádua ; De Menezes, Benjamin Rodrigues

Author_Institution

Univ. Fed. de Minas Gerais, Belo Horizonte, Brazil

fYear

2002

fDate

2002

Firstpage

44

Lastpage

48

Abstract

A variation of the well known Levenberg-Marquardt for training neural networks is presented in this work. The algorithm presented restricts the norm of the weights vector to a preestablished norm value and finds the minimum error solution for that norm value. A range of different norm solutions is generated and the best generalization solution is selected. The results show the efficiency of the algorithm in terms of convergence speed and generalization performance.

Keywords

convergence; generalisation (artificial intelligence); learning (artificial intelligence); neural nets; variable structure systems; Levenberg-Marquadt algorithm; convergence speed; generalization learning; generalization performance; minimum error solution; neural network training; sliding mode control; weight vector norm; Approximation algorithms; Computational complexity; Convergence; Equations; Error correction; Minimization methods; Neural networks; Optimization methods; Pareto optimization; Sliding mode control;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 2002. SBRN 2002. Proceedings. VII Brazilian Symposium on

Print_ISBN

0-7695-1709-9

Type

conf

DOI

10.1109/SBRN.2002.1181433

Filename

1181433