UDQL: Bridging The Gap between MSE Loss and The Optimal Value Function in Offline Reinforcement Learning
This addresses a key issue in offline RL for improving value estimation accuracy, though it appears incremental as it builds on existing methods.
The paper tackles the overestimation problem in offline reinforcement learning caused by using Mean Square Error (MSE) loss for value function estimation, proposing a novel Bellman underestimated operator and diffusion policy model that outperforms state-of-the-art algorithms on D4RL tasks.
The Mean Square Error (MSE) is commonly utilized to estimate the solution of the optimal value function in the vast majority of offline reinforcement learning (RL) models and has achieved outstanding performance. However, we find that its principle can lead to overestimation phenomenon for the value function. In this paper, we first theoretically analyze overestimation phenomenon led by MSE and provide the theoretical upper bound of the overestimated error. Furthermore, to address it, we propose a novel Bellman underestimated operator to counteract overestimation phenomenon and then prove its contraction characteristics. At last, we propose the offline RL algorithm based on underestimated operator and diffusion policy model. Extensive experimental results on D4RL tasks show that our method can outperform state-of-the-art offline RL algorithms, which demonstrates that our theoretical analysis and underestimation way are effective for offline RL tasks.