An incremental learning algorithm for functional approximation

This paper presents an incremental learning algorithm for feed-forward neural networks used as approximators of real world data. This algorithm allows neural networks of limited size to be obtained, providing better performances. The algorithm is compared to two of the main incremental algorithms (D...

Full description

Saved in:
Bibliographic Details
Published inAdvances in engineering software (1992) Vol. 40; no. 8; pp. 725 - 730
Main Authors Bahi, Jacques, Contassot-Vivier, Sylvain, Sauget, Marc
Format Journal Article
LanguageEnglish
Published Elsevier 2009
Subjects
Computer Science
Distributed, Parallel, and Cluster Computing
Incremental Learning
Neural networks
Online AccessGet full text

Cover

More Information
Summary:This paper presents an incremental learning algorithm for feed-forward neural networks used as approximators of real world data. This algorithm allows neural networks of limited size to be obtained, providing better performances. The algorithm is compared to two of the main incremental algorithms (Dunkin and cascade correlation) in the respective contexts of synthetic data and of real data consisting of radiation doses in homogeneous environments.
ISSN:0965-9978
DOI:10.1016/j.advengsoft.2008.12.018