Reinforcement Learning in Continuous State and Action Spaces door