Reorganization of Agent Networks with Reinforcement Learning Based on Communication Delay | IEEE Conference Publication | IEEE Xplore