Learning from Sparse Offline Datasets via Conservative Density Estimation
This addresses the challenge of learning effective policies from limited offline datasets for RL applications, though it appears incremental as it builds on existing methods like stationary distribution correction.
The paper tackles the problem of out-of-distribution extrapolation errors in offline reinforcement learning, especially in sparse reward or scarce data settings, by proposing Conservative Density Estimation (CDE), which achieves state-of-the-art performance on the D4RL benchmark.
Offline reinforcement learning (RL) offers a promising direction for learning policies from pre-collected datasets without requiring further interactions with the environment. However, existing methods struggle to handle out-of-distribution (OOD) extrapolation errors, especially in sparse reward or scarce data settings. In this paper, we propose a novel training algorithm called Conservative Density Estimation (CDE), which addresses this challenge by explicitly imposing constraints on the state-action occupancy stationary distribution. CDE overcomes the limitations of existing approaches, such as the stationary distribution correction method, by addressing the support mismatch issue in marginal importance sampling. Our method achieves state-of-the-art performance on the D4RL benchmark. Notably, CDE consistently outperforms baselines in challenging tasks with sparse rewards or insufficient data, demonstrating the advantages of our approach in addressing the extrapolation error problem in offline RL.