行動干渉状態の検出によるマルチエージェント強化学習法の改善

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: 2014/09/01

タイトル(英語): Improvement of Multiagent Reinforcement Learning Approach via Detection of Interference States

著者名: 許　海遅（千葉大学大学院工学研究科都市環境システムコース），荒井　幸代（千葉大学大学院工学研究科都市環境システムコース）

著者名(英語): Haichi XU (Division of Urban Environment Systems, Graduate School Faculty of Engineering, Chiba University), Sachiyo Arai (Division of Urban Environment Systems, Graduate School Faculty of Engineering, Chiba University)

キーワード: マルチエージェント学習，強化学習，エントロピー，状態表現　　Multiagent learning，Reinforcement learning，Entropy，State representation

要約(英語): In this paper, we propose a method to diminish the state space explosion problem of a multiagent reinforcement learning context, where each agent needs to observe other agents' states, and previous actions at each step of its learning process. However, both the number of state and action become exponential in the number of agents, leading to enormous amount of computation and very slow learning.In our method, the agent considers other agents' statuses only when they interfere with one another to reach their goals. Our idea is that each agent starts with its state space which does not include information of others'. Then, they automatically expand and refine their state space when agents detect interference. We adopt the information theory measure of entropy to detect the interference status where agents should take into account the other agents. We demonstrate the advantage of our method over the properties of global convergence in a time efficient manner.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.134 No.9 （2014） 特集Ⅰ：制御系設計における適応・学習・同定・モデリングの新展開　特集Ⅱ：インテリジェント・システム

本誌掲載ページ: 1310-1317 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/134/9/134_1310/_article/-char/ja/