Title of Paper:Convergence analysis of batch gradient algorithm for three classes of sigma-pi neural networks

Hits:

Date of Publication:2007-12-01

Journal:NEURAL PROCESSING LETTERS

Included Journals:SCIE、EI、Scopus

Volume:26

Issue:3

Page Number:177-189

ISSN No.:1370-4621

Key Words:convergence; sigma-Pi-Sigma neural networks; sigma-Sigma-Pi neural networks; sigma-Pi-Sigma-Pi neural networks; batch gradient algorithm; monotonicity

Abstract:Sigma-Pi (Sigma u-Sigma a) neural networks (SPNNs) are known to provide more powerful mapping capability than traditional feed-forward neural networks. A unified convergence analysis for the batch gradient algorithm for SPNN learning is presented, covering three classes of SPNNs: Sigma u-Sigma a-Sigma u, Sigma u-Sigma u-Sigma a and Sigma u-Sigma a-Sigma u-Sigma a. The monotonicity of the error function in the iteration is also guaranteed.