On Hard Exploration for Reinforcement Learning: A Case Study in Pommerman

Published in AAAI Conference on AI and Interactive Digital Entertainment, 2019

Recommended citation: Gao, C., Kartal, B., Hernandez-Leal, P. and Taylor, M.E., 2019, October. On hard exploration for reinforcement learning: A case study in pommerman. In Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (Vol. 15, No. 1, pp. 24-30). https://ojs.aaai.org/index.php/AIIDE/article/view/5220/5076

This paper discusses why exploration is hard for multi-agent RL in pommerman.

Download paper here