论文信息 - Coordination Structures Generated by Deep Reinforcement Learning in Distributed Task Executions

Coordination Structures Generated by Deep Reinforcement Learning in Distributed Task Executions

We investigate the coordination structures generated by deep Q-network (DQN) in a distributed task execution. Cooperation and coordination are the crucial issues in multi-agent systems, and very sophisticated design or learning is required in order to achieve effective structures or regimes of coordination. In this paper, we show the results that agents establish the division of labor in a bottom-up manner by determining their implicit responsible area when input structure for DQN is constituted by their own observation and absolute location.

Toshiharu Sugawara | Yuki Miyashita

[1] Mykel J. Kochenderfer,et al. Cooperative Multi-agent Control Using Deep Reinforcement Learning , 2017, AAMAS Workshops.

[2] Shane Legg,et al. Human-level control through deep reinforcement learning , 2015, Nature.

[3] Joel Z. Leibo,et al. Multi-agent Reinforcement Learning in Sequential Social Dilemmas , 2017, AAMAS.

[4] Rahul Savani,et al. Lenient Multi-Agent Deep Reinforcement Learning , 2017, AAMAS.

[5] Guillaume Lample,et al. Playing FPS Games with Deep Reinforcement Learning , 2016, AAAI.

[6] Gerhard Neumann,et al. Guided Deep Reinforcement Learning for Swarm Systems , 2017, ArXiv.

[7] Alex Graves,et al. Playing Atari with Deep Reinforcement Learning , 2013, ArXiv.