Presentation 2006-12-13
A Modified Solution Construction Method for Quadratic Assignment Problems
Naoto YOKOTA, Yoshihiko HORIO, Kazuyuki AIHARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) It is well known that the Hop field-type neural network does not always satisfy the constraints such as one neuron in each row and column, so that the network does not always give a feasible solution for a combinatorial optimization problem. To overcome this problem, a solution constructing method has been proposed to solve quadratic assignment problems (QAPs) with a chaotic neural network. However, when we apply the method to a mixed analog/digital chaotic neuron-computer hardware system, the solution constructing method does not show very good performance partly because of noise and circuit non-idealities, in particular, for large-scale QAPs. Therefore, we propose an improved solution constructing method. Through experiments with size-12 QAPs, we show superior performance of the proposed method to the original one. Moreover, simulations show that the proposed technique is effective even without noise and circuit non-idealities.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Chaotic Neuro-Computer / Neural Network Hardware / Quadratic Assignment Problems
Paper # NLP2006-94
Date of Issue

Conference Information
Committee NLP
Conference Date 2006/12/6(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 Modified Solution Construction Method for Quadratic Assignment Problems
Sub Title (in English)
Keyword(1) Chaotic Neuro-Computer
Keyword(2) Neural Network Hardware
Keyword(3) Quadratic Assignment Problems
1st Author's Name Naoto YOKOTA
1st Author's Affiliation Graduate School of Engineering, Tokyo Denki University()
2nd Author's Name Yoshihiko HORIO
2nd Author's Affiliation Graduate School of Engineering, Tokyo Denki University
3rd Author's Name Kazuyuki AIHARA
3rd Author's Affiliation Institute of Industrial Science, The University of Tokyo:Aihara Complexity Modelling Project, ERATO, JST
Date 2006-12-13
Paper # NLP2006-94
Volume (vol) vol.106
Number (no) 413
Page pp.pp.-
#Pages 6
Date of Issue