Distributed Reinforcement Learning with Self-Play in Parameterized Action Space | IEEE Conference Publication | IEEE Xplore