Presentation 2010-11-20
A Modular Neural Network for Parallel Computation
Daisuke SASAKI, Yoshihiro HAYAKAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Some of combinatorial optimization problems cause exponential increases of calculation time in terms of a problem size. In case of using a neural network solver, faster calculation is prospective, because each neuron is updated in essentially parallel. In fact, however, a neural network solver is performed by using the ordinary differential equation on a computer, so that a calculation speed is not enough yet. Hence, in this presentation we propose a modular neural network for the purpose of implementing parallel computation using by PC-Cluster. Each module is assigned to each processor and is calculated in parallel in our proposed modular neural network and we discuss the possibility of improvement of calculation by decreasing communication frequency between processors.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Combinatorial optimization problems / Neural Network / Parallel Computation
Paper # NLP2010-110
Date of Issue

Conference Information
Committee NLP
Conference Date 2010/11/12(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 Nonlinear Problems (NLP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Modular Neural Network for Parallel Computation
Sub Title (in English)
Keyword(1) Combinatorial optimization problems
Keyword(2) Neural Network
Keyword(3) Parallel Computation
1st Author's Name Daisuke SASAKI
1st Author's Affiliation Advanced Course of Information and Electronic, Sendai National College of Technology()
2nd Author's Name Yoshihiro HAYAKAWA
2nd Author's Affiliation Department of Information System, Sendai National College of Technology
Date 2010-11-20
Paper # NLP2010-110
Volume (vol) vol.110
Number (no) 299
Page pp.pp.-
#Pages 6
Date of Issue