Presentation 1993/11/24
An integrated paradigm for symbolism and connectionism
Hiroshi Tsukimoto,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Many researchers have pointed out that the integration of symbolism and connectionism,which are two big paradigms in AI and cognitive science,is necessary.This paper presents an integrated paradigm for classical logic,which is a fundamental part of symbolism and neural networks,which are fundamental parts of connectionism.The motif is to pay attention to the mathematical representation of the both.As a result,the space of multi-linear functions is obtained.The space becomes a Euclidean space by introducing an inner product.Thus,the functions in the space can be approximated.This new paradigm,the space of multi-linear functions,can be applied to several problems.An application to the structural analysis of neural networks is presented with an example of Exclusive OR.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) symbolism / connectionism / an integrated paradigm / neual networks / classical logic / multi-linear functions
Paper # NC93-53
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) An integrated paradigm for symbolism and connectionism
Sub Title (in English)
Keyword(1) symbolism
Keyword(2) connectionism
Keyword(3) an integrated paradigm
Keyword(4) neual networks
Keyword(5) classical logic
Keyword(6) multi-linear functions
1st Author's Name Hiroshi Tsukimoto
1st Author's Affiliation Toshiba()
Date 1993/11/24
Paper # NC93-53
Volume (vol) vol.93
Number (no) 341
Page pp.pp.-
#Pages 8
Date of Issue