連続行動空間への適用を考慮したSwitching強化学習

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2011/05/01

タイトル(英語): Switching Reinforcement Learning for Continuous Action Space

著者名: 永吉　雅人（新潟県立看護大学），村尾　元（神戸大学国際文化学部），玉置　久（神戸大学工学部）

著者名(英語): Masato Nagayoshi (Niigata College of Nursing), Hajime Murao (Faculty of Cross-Cultural Studies, Kobe University), Hisashi Tamaki (Faculty of Engineering, Kobe University)

キーワード: 強化学習，行動空間構成，状態空間構成，エントロピー，シミュレーション　　reinforcement learning，action space design，state space design，entropy，simulation

要約(英語): Reinforcement Learning (RL) attracts much attention as a technique of realizing computational intelligence such as adaptive and autonomous decentralized systems. In general, however, it is not easy to put RL into practical use. This difficulty includes a problem of designing a suitable action space of an agent, i.e., satisfying two requirements in trade-off: (i) to keep the characteristics (or structure) of an original search space as much as possible in order to seek strategies that lie close to the optimal, and (ii) to reduce the search space as much as possible in order to expedite the learning process.In order to design a suitable action space adaptively, we propose switching RL model to mimic a process of an infant's motor development in which gross motor skills develop before fine motor skills. Then, a method for switching controllers is constructed by introducing and referring to the “entropy”. Further, through computational experiments by using robot navigation problems with one and two-dimensional continuous action space, the validity of the proposed method has been confirmed.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.131 No.5 （2011） 特集：メタヒューリスティクスとその応用

本誌掲載ページ: 976-982 p

原稿種別: 論文／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejeiss/131/5/131_5_976/_article/-char/ja/