State Representation Learning With Adjacent State Consistency Loss for Deep Reinforcement Learning | IEEE Journals & Magazine | IEEE Xplore