Presentation 1994/10/13
Adaptive Conjugate Gradient Algorithm and its application to BP learning
Masayuki Gotoh, Nobuhiko Tawara,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) BP learning algorithm from Conjugate Gradient Method is effective method for high speed learning,However,in this method, problem which it depends on an initial setting of network parameter values should be mentioned.We,therefore,propose a learning algorithm with high accuracy based on Conjugate Gradient Method.The algorithm is based on "improved formiula" to create a conjugate direction and can search adapting parameter j.We call the proposed algorithm "Adaptive Conjugate Gradient algorithm.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Conjugate Gradient Method / Nonlinear programming / Backpropagation / Neural Networks
Paper # NC94-37
Date of Issue

Conference Information
Committee NC
Conference Date 1994/10/13(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) Adaptive Conjugate Gradient Algorithm and its application to BP learning
Sub Title (in English)
Keyword(1) Conjugate Gradient Method
Keyword(2) Nonlinear programming
Keyword(3) Backpropagation
Keyword(4) Neural Networks
1st Author's Name Masayuki Gotoh
1st Author's Affiliation Department of IE and Management,School of Science and Engineering, Waseda University()
2nd Author's Name Nobuhiko Tawara
2nd Author's Affiliation Department of Industrial Engineering,Faculty of Engineering, Musashi Institute of Technology
Date 1994/10/13
Paper # NC94-37
Volume (vol) vol.94
Number (no) 272
Page pp.pp.-
#Pages 8
Date of Issue