Magnified gradient function in adaptive learning: The MGFPROP algorithm

Sin Chun Ng, Chi Chung Cheung, Shu Hung Leung

Research output: Journal article publicationJournal articleAcademic researchpeer-review

1 Citation (Scopus)

Abstract

A new algorithm is proposed to solve the `flat spot' problem in back-propagation neural networks by magnifying the gradient function. Simulation results show that, in terms of the convergence rate and the percentage of global convergence, the new algorithm consistently outperforms other traditional methods.

Original languageEnglish
Pages (from-to)42-43
Number of pages2
JournalElectronics Letters
Volume37
Issue number1
DOIs
Publication statusPublished - 4 Jan 2001
Externally publishedYes

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Magnified gradient function in adaptive learning: The MGFPROP algorithm'. Together they form a unique fingerprint.

Cite this