Presentation 2003/3/10
III-posed Problems and Solutions in a Forward-propagation Learning Rule of Multi-layered Neural Networks
Yoshihiro OHAMA, Naohiro FUKUMURA, Yoji UNO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The existence of inverse models in motor control has been suggested. We have proposed that the inverse model can be acquired in an artificial multi-layered neural network using a forward-propagation learning rule. In this report, it is proposed that linear equations which may have ill-condition are solved for updating the weights of the forward-propagation learning rule. It was confirmed by computer simulation that the stability of learning depends on not only condition number but also change of singular value distribution of the equations.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) forward-propagation rule / inverse model / layered neural network / system identification / condition number
Paper # NC2002-146
Date of Issue

Conference Information
Committee NC
Conference Date 2003/3/10(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) III-posed Problems and Solutions in a Forward-propagation Learning Rule of Multi-layered Neural Networks
Sub Title (in English)
Keyword(1) forward-propagation rule
Keyword(2) inverse model
Keyword(3) layered neural network
Keyword(4) system identification
Keyword(5) condition number
1st Author's Name Yoshihiro OHAMA
1st Author's Affiliation Department of Information and Computer Sciences,Toyohashi University of Technology()
2nd Author's Name Naohiro FUKUMURA
2nd Author's Affiliation Department of Information and Computer Sciences,Toyohashi University of Technology
3rd Author's Name Yoji UNO
3rd Author's Affiliation Department of Information and Computer Sciences,Toyohashi University of Technology
Date 2003/3/10
Paper # NC2002-146
Volume (vol) vol.102
Number (no) 729
Page pp.pp.-
#Pages 6
Date of Issue