Reinforcement Learning Accelerated by Using State Transition Model With Robotic Applications

doi 10.1109/iros.2004.1389995