| Presentation | 2023-03-14 On Reward Distribution in Reinforcement Learning of Multi-Agent Surveillance Systems with Temporal Logic Specifications Keita Terashima, Koichi Kobayashi, Yuh Yamashita, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In multi-agent systems, it is important to design a reward distribution method based on the contribution of agents for efficient learning. In this paper, we propose a reward distribution method for a surveillance system based on a multi-agent reinforcement learning method using aggregators, where the control specification is described by a linear time-phase logic formula, which was previously proposed by the authors. In this method, the aggregator computes and distributes rewards according to the length of paths on the surveillance system. Finally, the performance is shown by numerical simulation with a surveillance problem as an example. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | multi-agent systems / reinforcement learning / linear temporal logic / aggregator / reward distribution problem / surveillance |
| Paper # | IT2022-81,ISEC2022-60,WBS2022-78,RCC2022-78 |
| Date of Issue | 2023-03-07 (IT, ISEC, WBS, RCC) |
| Conference Information | |
| Committee | RCC / ISEC / IT / WBS |
|---|---|
| Conference Date | 2023/3/14(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Shunichi Azuma(Nagoya Univ.) / Noboru Kunihiro(Tsukuba Univ.) / Tetsuya Kojima(Tokyo Kosen) / Takashi Shono(Wind River) |
| Vice Chair | Shunichi Azuma(Hokkaido Univ.) / Koji Ishii(Kagawa Univ.) / Junji Shikata(Yokohama National Univ.) / Goichiro Hanaoka(AIST) / Yasuyuki Nogami(Okayama Univ.) / Hiroyasu Ishikawa(Nihon Univ.) / Hideki Ochiai(Yokohama National Univ.) |
| Secretary | Shunichi Azuma(CRIEPI) / Koji Ishii(Ritsumeikan Univ.) / Junji Shikata(AIST) / Goichiro Hanaoka(Ibaraki Univ.) / Yasuyuki Nogami(Saitamai Univ.) / Hiroyasu Ishikawa(Nagaoka Univ. of Tech.) / Hideki Ochiai(Okayama Prefectural Univ.) |
| Assistant | SHAN LIN(NICT) / Ryosuke Adachi(Yamaguchi Univ.) / Yoshikazu Hanatani(Toshiba) / Takayuki Nozaki(Yamaguchi Univ.) / Sun Ran(Ibaraki Univ.) / Chen Na(NAIST) |
| Paper Information | |
| Registration To | Technical Committee on Reliable Communication and Control / Technical Committee on Information Security / Technical Committee on Information Theory / Technical Committee on Wideband System |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | On Reward Distribution in Reinforcement Learning of Multi-Agent Surveillance Systems with Temporal Logic Specifications |
| Sub Title (in English) | |
| Keyword(1) | multi-agent systems |
| Keyword(2) | reinforcement learning |
| Keyword(3) | linear temporal logic |
| Keyword(4) | aggregator |
| Keyword(5) | reward distribution problem |
| Keyword(6) | surveillance |
| 1st Author's Name | Keita Terashima |
| 1st Author's Affiliation | Hokkaido University(Hokkaido Univ.) |
| 2nd Author's Name | Koichi Kobayashi |
| 2nd Author's Affiliation | Hokkaido University(Hokkaido Univ.) |
| 3rd Author's Name | Yuh Yamashita |
| 3rd Author's Affiliation | Hokkaido University(Hokkaido Univ.) |
| Date | 2023-03-14 |
| Paper # | IT2022-81,ISEC2022-60,WBS2022-78,RCC2022-78 |
| Volume (vol) | vol.122 |
| Number (no) | IT-427,ISEC-428,WBS-429,RCC-430 |
| Page | pp.pp.86-90(IT), pp.86-90(ISEC), pp.86-90(WBS), pp.86-90(RCC), |
| #Pages | 5 |
| Date of Issue | 2023-03-07 (IT, ISEC, WBS, RCC) |