Randomized algorithms and PAC bounds for inverse reinforcement learning in continuous spaces

Open in new window