Scalable Reinforcement Learning Policies for Multi-Agent Control | IEEE Conference Publication | IEEE Xplore