Integrating Domain Knowledge for handling Limited Data in Offline RL
This addresses a critical bottleneck for real-world applications of offline RL where data is scarce, though it is incremental as it builds on existing regularization approaches.
The paper tackles the problem of offline reinforcement learning algorithms performing poorly with limited data confined to specific state regions, and it proposes a domain knowledge-based regularization technique that achieves an average performance increase of at least 27% compared to existing methods.
With the ability to learn from static datasets, Offline Reinforcement Learning (RL) emerges as a compelling avenue for real-world applications. However, state-of-the-art offline RL algorithms perform sub-optimally when confronted with limited data confined to specific regions within the state space. The performance degradation is attributed to the inability of offline RL algorithms to learn appropriate actions for rare or unseen observations. This paper proposes a novel domain knowledge-based regularization technique and adaptively refines the initial domain knowledge to considerably boost performance in limited data with partially omitted states. The key insight is that the regularization term mitigates erroneous actions for sparse samples and unobserved states covered by domain knowledge. Empirical evaluations on standard discrete environment datasets demonstrate a substantial average performance increase of at least 27% compared to existing offline RL algorithms operating on limited data.