Integrating Domain Knowledge for handling Limited Data in Offline RL
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913385177677824 |
|---|---|
| author | Gangopadhyay, Briti Wang, Zhao Yeh, Jia-Fong Takamatsu, Shingo |
| author_facet | Gangopadhyay, Briti Wang, Zhao Yeh, Jia-Fong Takamatsu, Shingo |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_07041 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Integrating Domain Knowledge for handling Limited Data in Offline RL Gangopadhyay, Briti Wang, Zhao Yeh, Jia-Fong Takamatsu, Shingo Machine Learning Artificial Intelligence 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. |
| title | Integrating Domain Knowledge for handling Limited Data in Offline RL |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2406.07041 |