Presentation 2003/6/20
A self-organizing model of distributed representation with various kind of feature detectors
Hidemasa YOSHIDA, Toshio AOYAGI, Koji KURATA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A novel type of self-organizing feature map model for multiple dimensional stimulus space is presented and studied numerically. The essential algorithm is based on competitive neural networks in which we assume that multiple neurons are selected and their connections from an input layer are modified by Hebbian rule. In our new model information is represented by simultaneous activation of different kind of feature detectors. Moreover, as a result of learning, the neural array is divided into two areas detecting two different kind of features.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) self-organization / distributed representation / winner-share-all / orthogonalization
Paper # NC2003-19
Date of Issue

Conference Information
Committee NC
Conference Date 2003/6/20(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 self-organizing model of distributed representation with various kind of feature detectors
Sub Title (in English)
Keyword(1) self-organization
Keyword(2) distributed representation
Keyword(3) winner-share-all
Keyword(4) orthogonalization
1st Author's Name Hidemasa YOSHIDA
1st Author's Affiliation Department of Applied Analysis and Complex Dynamical Systems, Graduate School of Infomatics, Kyoto University()
2nd Author's Name Toshio AOYAGI
2nd Author's Affiliation Department of Applied Analysis and Complex Dynamical Systems, Graduate School of Infomatics, Kyoto University
3rd Author's Name Koji KURATA
3rd Author's Affiliation Department of Mechanical System Engineering, Faculty of Engineering, University of the Ryukyus
Date 2003/6/20
Paper # NC2003-19
Volume (vol) vol.103
Number (no) 153
Page pp.pp.-
#Pages 5
Date of Issue