Presentation 1994/11/29
Neuro-Automata and Soft Codes
Takeo Ikai, Kazuyuki Hatori,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Neuro-automata(NA's)are realized by means of second-order recurrent neural networks and behave as finite automata(FA's) according to input sequences over {0,1} which are codes of input symbols.We have studied,in this paper,the behavior of NA's for real valued input(soft codes)sequences over[0,1]which represent subsymbolic analog information.As a result,the followings are obtained.Symbols are corresponding to subsymbolic equivalence classes which cause the same state transitions,and an NA accepts subsymbolic sequences corresponding to regular sequences accepted by its extended FA.Subsymbol automata are then formulated and the framework of automata theory are extended.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) neuro-automate / soft codes / subsymbols / symbols as subsymbol equivalence classes / subsymbol automata
Paper # NC94-42
Date of Issue

Conference Information
Committee NC
Conference Date 1994/11/29(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) Neuro-Automata and Soft Codes
Sub Title (in English)
Keyword(1) neuro-automate
Keyword(2) soft codes
Keyword(3) subsymbols
Keyword(4) symbols as subsymbol equivalence classes
Keyword(5) subsymbol automata
1st Author's Name Takeo Ikai
1st Author's Affiliation Faculty of Engineering,University of Osaka Prefecture()
2nd Author's Name Kazuyuki Hatori
2nd Author's Affiliation Faculty of Engineering,University of Osaka Prefecture
Date 1994/11/29
Paper # NC94-42
Volume (vol) vol.94
Number (no) 377
Page pp.pp.-
#Pages 8
Date of Issue