Metraipolis › archive › dc_6969ef417391

Improving Generalization and Robustness in Offline Reinforcement Learning via Boundary-Aware Data Augmentation — arXiv:2609.20300

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.20300
License (as archived)
CC0-1.0
Archived
2026-10-10T04:09:00.169Z
SHA-256 of the archived text
d175c10cc087a2e31108490e78a2418a1955c7f985bcfb3fd77d0bca96fa9cbf
Stable URL of this page
https://metraipolis.com/documento/dc_6969ef417391

How to cite

Improving Generalization and Robustness in Offline Reinforcement Learning via Boundary-Aware Data Augmentation — arXiv:2609.20300 — arXiv (en), https://arxiv.org/abs/2609.20300 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6969ef417391 (Metraipolis archive dc_6969ef417391, sha256 d175c10cc087a2e31108490e78a2418a1955c7f985bcfb3fd77d0bca96fa9cbf)
BibTeX
@misc{dc6969ef417391,
  title = {Improving Generalization and Robustness in Offline Reinforcement Learning via Boundary-Aware Data Augmentation — arXiv:2609.20300},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.20300},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6969ef417391; sha256 d175c10cc087a2e31108490e78a2418a1955c7f985bcfb3fd77d0bca96fa9cbf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.20300. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Current offline reinforcement learning (ORL) algorithms tend to overfit the training dataset and exhibit poor in-distribution generalization and robustness performance when deployed to real environments, thus compromising their effectiveness. Existing methods typically enhance in-distribution generalization and robustness by leveraging regularization techniques widely used in computer vision. However, due to the high sensitivity of low-level physical signals to distributional shifts, these methods still suffer from notable limitations in in-distribution generalization and robustness, making it difficult to achieve stable performance in complex environments. To address this issue, we theoretically analyze the error bounds of the behavior policy and action-value function trained with random episode interpolation, revealing that the error scales positively correlated with the distance between states. Based on this insight, we propose a method called $\bf{B}$oundary-$\bf{A}$ware $\bf{D}$ata $\bf{A}$ugmentation (BADA), which leverages neighboring states to construct interpolation boundaries, enabling the generation of synthetic data that more faithfully preserves the original data distribution. We first conduct qualitative studies in a toy environment, showing that BADA generates mixed samples that preserve desirable policy smoothness while accurately reconstructing multimodal value distributions. Extensive experiments on limited offline datasets further demonstrate that BADA attains state-of-the-art performance across diverse benchmarks.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.20300 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T03:48:18.352Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).