Accelerating Robot Reinforcement Learning with Samples of Different Simulation Precision | IEEE Conference Publication | IEEE Xplore