Presentation 1993/9/22
Parallel Learning by Threshold Fluctuation for Multi-layered Perceptron.
Kazuhisa Watanabe, Hajime Masuda, Mitsuaki Mitani, Takafumi Ohhori,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a general-purpose parallel learning rule different from the back-propagation algorithm.The proposed rule can be applied to feedforward neural-networks.Each unit has a threshold value with small and independent flactuation.Analysis clarifies that each weight can be modified separately by leraning rule using output average of the front unit and correlation of the network output fluctuation and threshold fluctuation within the rear unit. Similations of XOR and random mapping problems have demonstrated good performance.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) threshold / fluctuation / correlation / average / parallel learning / feedforward neural network
Paper # NC93-39
Date of Issue

Conference Information
Committee NC
Conference Date 1993/9/22(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) Parallel Learning by Threshold Fluctuation for Multi-layered Perceptron.
Sub Title (in English)
Keyword(1) threshold
Keyword(2) fluctuation
Keyword(3) correlation
Keyword(4) average
Keyword(5) parallel learning
Keyword(6) feedforward neural network
1st Author's Name Kazuhisa Watanabe
1st Author's Affiliation Department of Electrical Engineering,Hokkaido Institute of Technology()
2nd Author's Name Hajime Masuda
2nd Author's Affiliation Department of Electrical Engineering,Hokkaido Institute of Technology
3rd Author's Name Mitsuaki Mitani
3rd Author's Affiliation Department of Electrical Engineering,Hokkaido Institute of Technology
4th Author's Name Takafumi Ohhori
4th Author's Affiliation Department of Electrical Engineering,Hokkaido Institute of Technology
Date 1993/9/22
Paper # NC93-39
Volume (vol) vol.93
Number (no) 247
Page pp.pp.-
#Pages 7
Date of Issue