Mini-max initialization for function approximation

Document Type

Article

Publication Date

3-1-2004

Abstract

Neural networks have been successfully applied to various pattern recognition and function approximation problems. However, the training process remains a time-consuming procedure that often gets stuck in a local minimum. The optimum network size and topology are usually unknown. In this paper, we formulate the concept of extrema equivalence for estimating the complexity of a function. Based on this formulation, the optimal network size and topology can be selected according to the number of extrema. Mini-max initialization method is then proposed to select the initial values of the weights for the network that is proven to greatly speed up training. The superior performance of our method in terms of convergence and generalization has been substantiated by experimental results. © 2003 Elsevier B.V. All rights reserved.

Identifier

1542471417 (Scopus)

Publication Title

Neurocomputing

External Full Text Location

https://doi.org/10.1016/j.neucom.2003.10.014

ISSN

09252312

First Page

389

Last Page

409

Issue

1-4

Volume

57

Fund Ref

State of New Jersey Commission on Science and Technology

This document is currently not available here.

Share

COinS