Adversarial Policy Learning in Two-player Competitive Games

                  Wenbo Guo 1 Xian Wu 1 Sui Huang 2 Xinyu Xing 1

             Abstract                2020), we argue that attacks developed under this assump-
                                 tion are not practical. For example, given a master agent
    In a two-player deep reinforcement learning task,
                                 deployed in an online video game platform playing with
    recent work shows an attacker could learn ...