Presentation 1999/11/26
A Study on Initialization for Structural Learning with Forgetting
Hiroshi SHIRATSUCHI, Hiromu GOTANDA, Masaaki SHIMADA, Katsuhiro INOUE, Kousuke KUMAMARU,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper studies how our proposed initialization effects the structure determination of neural networks by structural learning with forgetting. The proposed initialization consists of two steps: weights of hidden units are initialized so that their hyperplanes should pass through the center of input pattern set, and those of output units are initialized to zero. From simulation result performed on Iris problem, it was confirmed that the initialization gives better network structure and higher generalization ability.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Intitialization / Strctual Learning with Forgetting / Structure Determination
Paper # NC99-59
Date of Issue

Conference Information
Committee NC
Conference Date 1999/11/26(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) A Study on Initialization for Structural Learning with Forgetting
Sub Title (in English)
Keyword(1) Intitialization
Keyword(2) Strctual Learning with Forgetting
Keyword(3) Structure Determination
1st Author's Name Hiroshi SHIRATSUCHI
1st Author's Affiliation University of the Ryukyus()
2nd Author's Name Hiromu GOTANDA
2nd Author's Affiliation Kinki University in Kyushu
3rd Author's Name Masaaki SHIMADA
3rd Author's Affiliation Kyushu Institute of Technology
4th Author's Name Katsuhiro INOUE
4th Author's Affiliation Kyushu Institute of Technology
5th Author's Name Kousuke KUMAMARU
5th Author's Affiliation Kyushu Institute of Technology
Date 1999/11/26
Paper # NC99-59
Volume (vol) vol.99
Number (no) 473
Page pp.pp.-
#Pages 7
Date of Issue