A Study on Multi-point Search Combinatorial Optimization Method Based on Big Valley Structure

カテゴリ: 部門大会

論文No: SS4-3

グループ名: 【C】平成27年電気学会電子・情報・システム部門大会講演論文集

発行日: 2015/08/27

タイトル(英語): A Study on Multi-point Search Combinatorial Optimization Method Based on Big Valley Structure

著者名: Masahide Morita(Tokyo Metropolitan University),Hiroki Ochiai(Tokyo Metropolitan University),Kenichi Tamura(Tokyo Metropolitan University),Keiichiro Yasuda(Tokyo Metropolitan University)

著者名(英語): Masahide Morita(Tokyo Metropolitan University),Hiroki Ochiai(Tokyo Metropolitan University),Kenichi Tamura(Tokyo Metropolitan University),Keiichirou Yasuda(Tokyo Metropolitan University)

キーワード: Meta-heuristics|Combinatorial Optimization|Proximate Optimality Principle|Big Valley Structure|Meta-heuristics|Combinatorial Optimization|Proximate Optimality Principle|Big Valley Structure

要約(日本語): In recent years, meta-heuristics, which is practical combinatorial optimization method, has been noted. We understand that searches of meta-heuristics share the same structure, which is the Proximate Optimality Principle (POP). On the other hand, we interpret the big valley structure as ``a sign of POP in the solution-evaluation value space,'' and focus on the fact that the degree of its establishment is different in each problem. In addition, we note that it can be expected that it is possible to quantitatively evaluate the degree by using the correlation correlation .On the basis of the above, we propose a new multi-point combinatorial optimization method. The performance of the proposed method is verified through simulations by employing two types of typical benchmark problems.

PDFファイルサイズ: 189 Kバイト