1 paper · 1 filter
Dylan Klein, Akansel Cosgun
We investigate the effect of using human demonstration data in the replay buffer for Deep Reinforcement Learning. We use a policy gradient method with a modified experience replay…