Presentation 1996/3/18
Analysis of Continuous-Time Neuro-Window Networks
Hideki Kakeya, Toshiki Kindo, Masato Okada, Yoichi Okabe,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In the present paper the authors extend the neuro-window networks, which realize hierarchical concept formation and association, to continuous dynamics. In the continuous model, the basin of attraction expands. Also the low-level concept patterns are recalled on the way of high-level concept association. These phenomena are elucidated from the viewpoint of linear algebra and the analysis on the differential equations of the finite system.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) associative memory / neuro-window network / continuous dynamics / nonmonotonic neuron / concept formation / hierarchical concept / autocorrelation / geometrical analysis / finite system
Paper # NC-95-131
Date of Issue

Conference Information
Committee NC
Conference Date 1996/3/18(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 of Continuous-Time Neuro-Window Networks
Sub Title (in English)
Keyword(1) associative memory
Keyword(2) neuro-window network
Keyword(3) continuous dynamics
Keyword(4) nonmonotonic neuron
Keyword(5) concept formation
Keyword(6) hierarchical concept
Keyword(7) autocorrelation
Keyword(8) geometrical analysis
Keyword(9) finite system
1st Author's Name Hideki Kakeya
1st Author's Affiliation RCAST, University of Tokyo()
2nd Author's Name Toshiki Kindo
2nd Author's Affiliation Research Development Corporation of Japan
3rd Author's Name Masato Okada
3rd Author's Affiliation Dept. of Biophysical Engineering, Osaka Univ.
4th Author's Name Yoichi Okabe
4th Author's Affiliation RCAST, University of Tokyo
Date 1996/3/18
Paper # NC-95-131
Volume (vol) vol.95
Number (no) 598
Page pp.pp.-
#Pages 8
Date of Issue