複数の状態行動価値表を用いたR学習の高速化  [in Japanese] R-learning with Multiple State-action Value Tables  [in Japanese]

Access this Article

Search this Article

Author(s)

    • 櫻井 彰人 SAKURAI Akito
    • 慶應義塾大学 理工学部 管理工学科 Department of Administration Engineering, Faculty of Science and Technology, Keio University
    • 藤波 努 [他] FUJINAMI Tsutomu
    • 北陸先端科学技術大学院大学 知識科学研究科 School of Knowledge Science, Japan Advanced Institute of Science and Technology
    • 國藤 進 KUNIFUJI Susumu
    • 北陸先端科学技術大学院大学 知識科学研究科 School of Knowledge Science, Japan Advanced Institute of Science and Technology

Abstract

We propose a method to improve the performance of R-learning, a reinforcement learning algorithm, by using multiple state-action value tables. Unlike Q- or Sarsa learning, R-learning learns a policy to maximize undiscounted rewards. Multiple state-action value tables cause substantial explorations as needed and make R-learnings to work well. Efficiency of the proposed method is verified through experiments in simulation environment.

Journal

  • IEEJ Transactions on Electronics, Information and Systems

    IEEJ Transactions on Electronics, Information and Systems 126(1), 72-82, 2006-01-01

    The Institute of Electrical Engineers of Japan

References:  9

Codes

  • NII Article ID (NAID)
    10016922097
  • NII NACSIS-CAT ID (NCID)
    AN10065950
  • Text Lang
    JPN
  • Article Type
    ART
  • ISSN
    03854221
  • NDL Article ID
    7786275
  • NDL Source Classification
    ZN31(科学技術--電気工学・電気機械工業)
  • NDL Call No.
    Z16-795
  • Data Source
    CJP  NDL  J-STAGE 
Page Top