Presentation 2018-03-12
Learning in Two-Player Matrix Games by Policy Gradient Lagging Anchor
Shiyao Ding, Toshimitsu Ushio,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a novel multi-agent reinforcement learning (MARL) algorithm which is called a policy gra-dient lagging anchor (PGLA) algorithm. Then, we consider 2 two-player matrix games as illustrative examples. Andit is shown by simulation that behaviors of the games using the PGLA algorithm can converge to Nash equilibriain both pure and mixed policies.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Reinforcement Learning / Policy Gradient / Multi-Agent Systems / Matrix Game
Paper # MSS2017-79
Date of Issue 2018-03-05 (MSS)

Conference Information
Committee MSS / NLP
Conference Date 2018/3/12(3days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Morikazu Nakamura(Univ. of Ryukyus) / Masaharu Adachi(Tokyo Denki Univ.)
Vice Chair Shigemasa Takai(Osaka Univ.) / Norikazu Takahashi(Okayama Univ.)
Secretary Shigemasa Takai(Toshiba) / Norikazu Takahashi(Osaka Univ.)
Assistant Hideki Kinjo(Okinawa Univ.) / Toshihiro Tachibana(Shonan Inst. of Tech.) / Masayuki Kimura(Kyoto Univ.)

Paper Information
Registration To Technical Committee on Mathematical Systems Science and its applications / Technical Committee on Nonlinear Problems
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Learning in Two-Player Matrix Games by Policy Gradient Lagging Anchor
Sub Title (in English)
Keyword(1) Reinforcement Learning
Keyword(2) Policy Gradient
Keyword(3) Multi-Agent Systems
Keyword(4) Matrix Game
1st Author's Name Shiyao Ding
1st Author's Affiliation Osaka University(Osaka Univ.)
2nd Author's Name Toshimitsu Ushio
2nd Author's Affiliation Osaka University(Osaka Univ.)
Date 2018-03-12
Paper # MSS2017-79
Volume (vol) vol.117
Number (no) MSS-506
Page pp.pp.11-14(MSS),
#Pages 4
Date of Issue 2018-03-05 (MSS)