Presentation 1993/11/24
Construction Multi-Module Neural Networks and Recognition of English words
Hiroshi Inaba, Katsunori Tanabe,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Introducing a module neural network(MNN)as a McCullough-Pitts model dynamical neural network having a special structure,and constructing a multi-module neural dynamical network(MMNN)by mutually connecting several modules,more complicate associative memories are proposed.This MMNN is constructed so as to keep the equilibrium solution of each MNN and to assign the desired equilibrium solutions for the entire MMNN.A method is proposed for constructiing the connection matrix of such a MMNN so that the given patterns to be memorized are assigned to the equilibrium solutions of the MMNN.As an application,computer simulation of recognition of English words is performed.The simulation result indicates that all the given patterns are assigned as stable equilibrium solutions of the MMNN.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) McCullough-Pitts Model / Multi-Module Neural Network / Equilibrium solution / Generalized Inverse
Paper # NC93-46
Date of Issue

Conference Information
Committee NC
Conference Date 1993/11/24(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) Construction Multi-Module Neural Networks and Recognition of English words
Sub Title (in English)
Keyword(1) McCullough-Pitts Model
Keyword(2) Multi-Module Neural Network
Keyword(3) Equilibrium solution
Keyword(4) Generalized Inverse
1st Author's Name Hiroshi Inaba
1st Author's Affiliation Department of Information Sciences,Tokyo Denki University()
2nd Author's Name Katsunori Tanabe
2nd Author's Affiliation Department of Information Sciences,Tokyo Denki University
Date 1993/11/24
Paper # NC93-46
Volume (vol) vol.93
Number (no) 341
Page pp.pp.-
#Pages 8
Date of Issue