Presentation 1996/10/28
Analysis and Improvement of a Spurious-Free Associative Memory System
Atsushi Ohsaki, Kazushi Ikeda, Kenji Nakayama,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The existence of spurious memories is a serious problem in associative memory systems based on recurrent neural networks. A spurious-free associative memory system which consists of hysteresis neuron and pseudo-inverse matrix weights has been proposed. This system, however, has a problem that the basins of stored patterns small. In this paper, we analyze the system, improve it to have large basins, and show the effect experiment ally.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) associative memory / spurious memory / hysteresis neuron / pseudo-inverse matrix model / basin
Paper # NC96-41
Date of Issue

Conference Information
Committee NC
Conference Date 1996/10/28(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) Analysis and Improvement of a Spurious-Free Associative Memory System
Sub Title (in English)
Keyword(1) associative memory
Keyword(2) spurious memory
Keyword(3) hysteresis neuron
Keyword(4) pseudo-inverse matrix model
Keyword(5) basin
1st Author's Name Atsushi Ohsaki
1st Author's Affiliation Dept. of Electrical and Computer Eng., Faculty of Tech:, Kanazawa Univ.()
2nd Author's Name Kazushi Ikeda
2nd Author's Affiliation Dept. of Electrical and Computer Eng., Faculty of Tech:, Kanazawa Univ.
3rd Author's Name Kenji Nakayama
3rd Author's Affiliation Dept. of Electrical and Computer Eng., Faculty of Tech:, Kanazawa Univ.
Date 1996/10/28
Paper # NC96-41
Volume (vol) vol.96
Number (no) 331
Page pp.pp.-
#Pages 8
Date of Issue