このエントリーをはてなブックマークに追加
ID 19716
Eprint ID
19716
フルテキストURL
著者
Karatsu Naoya Tokyo Institute of Technology
Nagata Yuichi Tokyo Institute of Technology
Ono Isao Tokyo Institute of Technology
Kobayashi Shigenobu Tokyo Institute of Technology
抄録
When attempting to optimize a function where exists several big-valley structures, conventional GAs often fail to find the global optimum. Innately Split Model (ISM) is a framework of GAs, which is designed to avoid this phenomenon called UV-Phenomenon. However, ISM doesn't care about previously-searched areas by the past populations. Thus, it is possible that populations of ISM waste evaluation cost for redundant searches reaching previously-found optima. In this paper, we introduce Extended ISM (EISM) that uses search information of past populations as trap to suppress overlapping searches. To show performance of EISM, we apply it to some test functions, and analyze the behavior.
発行日
2009-11-12
出版物タイトル
Proceedings : Fifth International Workshop on Computational Intelligence & Applications
2009巻
1号
出版者
IEEE SMC Hiroshima Chapter
開始ページ
284
終了ページ
289
ISSN
1883-3977
NCID
BB00577064
資料タイプ
会議発表論文
言語
English
OAI-PMH Set
岡山大学
著作権者
IEEE SMC Hiroshima Chapter
イベント
5th International Workshop on Computational Intelligence & Applications IEEE SMC Hiroshima Chapter : IWCIA 2009
イベント地
東広島市
イベント地の別言語
Higashi-Hiroshima City
論文のバージョン
publisher
査読
有り
Sort Key
53
Eprints Journal Name
IWCIA