Parameter Sharing Deep Deterministic Policy Gradient for Cooperative Multi-agent Reinforcement Learning
arXiv: Artificial Intelligence, Volume abs/1710.00336, 2017.
Deep reinforcement learning for multi-agent cooperation and competition has been a hot topic recently. This paper focuses on cooperative multi-agent problem based on actor-critic methods under local observations settings. Multi agent deep deterministic policy gradient obtained state of art results for some multi-agent games, whereas, it c...More
Full Text (Upload PDF)
PPT (Upload PPT)