A Deeper Look at Experience Replay
This addresses a hyper-parameter tuning problem for deep reinforcement learning practitioners, offering an incremental improvement.
The paper investigates the impact of experience replay in deep reinforcement learning, finding that large replay buffers can significantly hurt performance, and proposes a simple O(1) method to mitigate this issue, validated in grid world and Atari games.
Recently experience replay is widely used in various deep reinforcement learning (RL) algorithms, in this paper we rethink the utility of experience replay. It introduces a new hyper-parameter, the memory buffer size, which needs carefully tuning. However unfortunately the importance of this new hyper-parameter has been underestimated in the community for a long time. In this paper we did a systematic empirical study of experience replay under various function representations. We showcase that a large replay buffer can significantly hurt the performance. Moreover, we propose a simple O(1) method to remedy the negative influence of a large replay buffer. We showcase its utility in both simple grid world and challenging domains like Atari games.