Leveraging Demonstrations for Deep Reinforcement Learning on Robotics Problems with Sparse RewardsMartin Riedmiller, Todd Hester, Bilal Piot et al.|arXiv (Cornell University)|2017Cited by 510