Presentation 2016-12-12
The Relationship between Modular Neural Network and Noise Effect
Kou Muraoka, Rui Yoshida, Yoshihiro Hayakawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Some of combinatorial optimization problems have problems that the number of solutions to be searched increases with the problem size and the computation cost becomes enormous. As a method to solve these problems, a method using a Neural Network (NN) has been proposed. However, in normal simulation, the parallelism of NN is lost. For this reason, a Modular Neural Network (MNN) has been proposed in which parallel computation is directed and computation is performed among plurality of processors. It is shown that this method improves the solution searching ability by adding fluctuation to the com-munication interval time between processors. In this paper, we investigate the performance of MNN which fluctuates the communication interval time for the 8-Queens problem and the reason for improving the correct answer rate in the viewpoint of noise effect.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Combinatorial optimization problems / Neural Network / Noise effect
Paper # NLP2016-91
Date of Issue 2016-12-05 (NLP)

Conference Information
Committee NLP
Conference Date 2016/12/12(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Chukyo Univ.
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Hisato Fujisaka(Hiroshima City Univ.)
Vice Chair Masaharu Adachi(Tokyo Denki Univ.)
Secretary Masaharu Adachi(Konan Univ.)
Assistant Hiroyuki Asahara(Okayama Univ. of Science) / Toshihiro Tachibana(Shonan Inst. of Tech.)

Paper Information
Registration To Technical Committee on Nonlinear Problems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) The Relationship between Modular Neural Network and Noise Effect
Sub Title (in English)
Keyword(1) Combinatorial optimization problems
Keyword(2) Neural Network
Keyword(3) Noise effect
1st Author's Name Kou Muraoka
1st Author's Affiliation National Institute of Technology, Sendai College(NIT, Sendai)
2nd Author's Name Rui Yoshida
2nd Author's Affiliation National Institute of Technology, Sendai College(NIT, Sendai)
3rd Author's Name Yoshihiro Hayakawa
3rd Author's Affiliation National Institute of Technology, Sendai College(NIT, Sendai)
Date 2016-12-12
Paper # NLP2016-91
Volume (vol) vol.116
Number (no) NLP-353
Page pp.pp.39-44(NLP),
#Pages 6
Date of Issue 2016-12-05 (NLP)