A Deep Reinforcement Learning Model for a Two-Layer Scheduling Policy in Urban Public Resources | IEEE Journals & Magazine | IEEE Xplore