Presentation 2014-07-21
Solving Ability of Lin-Kernighan Method Driven by Chaotic Dynamics for Traveling Salesman Problems
Takahiro MITSUOKA, Mikio HASEGAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Effectiveness of chaos for optimization has been shown by many previous researches. In this paper, a chaotic search based on the 2-opt for the Traveling Salesman Problem (TSP) is extended to the k-opt based version. A chaotic search based on the adaptive k-opt, whose k is adaptively changed in similar way to the Lin-Kernighan method, is proposed. Our proposed method is based on the tabu search. First, a conventional tabu search is implemented on a neural network model, which has refractory effects on each neuron. The tabu search neural network can be simply extended to chaotic neural network, by replacing the output function with analog sigmoidal function. By numerical simulations, it is shown that the proposed method solves good near optimum solutions, which is about 1% different from the best known solutions, with only 5000 iterations even for an 11849-city TSP.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Chaos / Combinatorial Optimization / Neural Networks / Lin-Kernighan Method / Tabu Search
Paper # NLP2014-35
Date of Issue

Conference Information
Committee NLP
Conference Date 2014/7/14(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) Solving Ability of Lin-Kernighan Method Driven by Chaotic Dynamics for Traveling Salesman Problems
Sub Title (in English)
Keyword(1) Chaos
Keyword(2) Combinatorial Optimization
Keyword(3) Neural Networks
Keyword(4) Lin-Kernighan Method
Keyword(5) Tabu Search
1st Author's Name Takahiro MITSUOKA
1st Author's Affiliation Faculty of Engineering, Tokyo University of Science()
2nd Author's Name Mikio HASEGAWA
2nd Author's Affiliation Faculty of Engineering, Tokyo University of Science
Date 2014-07-21
Paper # NLP2014-35
Volume (vol) vol.114
Number (no) 145
Page pp.pp.-
#Pages 4
Date of Issue