Presentation 1994/7/25
Recalling method useing energy of network on Perceptron type network.
Tadashi Aizawa, Akira Hyogo, Keitaro Sekine,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Method of recalling paturn on the Hopfeild type network is realized by decreasing energy of network.This method can remenber a closed paturn to input pattern for energy of network.While Hopfield type network can′t learn patterns whose pattern vectors d on′t cross at righl angle.On the other hand Perceptron type networ k can learn any pattern.So we propose to use recalling way which use energy of network on the Perceptron type network.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Recalling / Perceptron type network / Energy of network
Paper # NC94-24
Date of Issue

Conference Information
Committee NC
Conference Date 1994/7/25(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) Recalling method useing energy of network on Perceptron type network.
Sub Title (in English)
Keyword(1) Recalling
Keyword(2) Perceptron type network
Keyword(3) Energy of network
1st Author's Name Tadashi Aizawa
1st Author's Affiliation Faculty of Science and Technology,Science University of Tokyo()
2nd Author's Name Akira Hyogo
2nd Author's Affiliation Faculty of Science and Technology,Science University of Tokyo
3rd Author's Name Keitaro Sekine
3rd Author's Affiliation Faculty of Science and Technology,Science University of Tokyo
Date 1994/7/25
Paper # NC94-24
Volume (vol) vol.94
Number (no) 182
Page pp.pp.-
#Pages 7
Date of Issue