Partitioning input space for reinforcement learning for control | IEEE Conference Publication | IEEE Xplore