论文信息 - Multiplayer Support for the Arcade Learning Environment

Multiplayer Support for the Arcade Learning Environment

The Arcade Learning Environment ("ALE") is a widely used library in the reinforcement learning community that allows easy program-matic interfacing with Atari 2600 games, via the Stella emulator. We introduce a publicly available extension to the ALE that extends its support to multiplayer games and game modes. This interface is additionally integrated with PettingZoo to allow for a simple Gym-like interface in Python to interact with these games.

Benjamin Black | Justin K. Terry | J. K. Terry | Benjamin Black

[1] Benjamin Black,et al. SuperSuit: Simple Microwrappers for Reinforcement Learning Environments , 2020, ArXiv.

[2] Yi Wu,et al. Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments , 2017, NIPS.

[3] Tom Schaul,et al. Rainbow: Combining Improvements in Deep Reinforcement Learning , 2017, AAAI.

[4] Demis Hassabis,et al. Mastering the game of Go without human knowledge , 2017, Nature.

[5] Dinesh Manocha,et al. Parameter Sharing is Surprisingly Useful for Multi-Agent Deep Reinforcement Learning. , 2020 .

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

[7] Marlos C. Machado,et al. Revisiting the Arcade Learning Environment: Evaluation Protocols and Open Problems for General Agents , 2017, J. Artif. Intell. Res..

[8] Marc G. Bellemare,et al. The Arcade Learning Environment: An Evaluation Platform for General Agents , 2012, J. Artif. Intell. Res..

[9] Wojciech Zaremba,et al. OpenAI Gym , 2016, ArXiv.

[10] Kenneth O. Stanley,et al. Go-Explore: a New Approach for Hard-Exploration Problems , 2019, ArXiv.

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