Bridging Domain Gaps with Target-Aligned Generation for Offline Reinforcement Learning

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Kim, Minung, Kim, Jeongmo, Choi, Gwanwoo, Han, Seungyul
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916008647720960
author Kim, Minung
Kim, Jeongmo
Choi, Gwanwoo
Han, Seungyul
author_facet Kim, Minung
Kim, Jeongmo
Choi, Gwanwoo
Han, Seungyul
contents Cross-domain offline reinforcement learning aims to adapt a policy from a source domain to a target domain using only pre-collected datasets, where environment dynamics may differ. A key challenge is to leverage source data while reducing distributional mismatch, particularly when the target dataset is extremely limited. To address this, we propose Target-aligned Coverage Expansion (TCE), a framework that decides how source data should be used, either by directly incorporating target-near transitions or by expanding state coverage through target-aligned generation, guided by theoretical analysis. TCE builds on a dual score-based generative model to synthesize target-consistent transitions over an expanded state region. Extensive experiments across diverse cross-domain environments show that TCE consistently outperforms state-of-the-art cross-domain offline RL baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2605_13054
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Bridging Domain Gaps with Target-Aligned Generation for Offline Reinforcement Learning
Kim, Minung
Kim, Jeongmo
Choi, Gwanwoo
Han, Seungyul
Machine Learning
Artificial Intelligence
Cross-domain offline reinforcement learning aims to adapt a policy from a source domain to a target domain using only pre-collected datasets, where environment dynamics may differ. A key challenge is to leverage source data while reducing distributional mismatch, particularly when the target dataset is extremely limited. To address this, we propose Target-aligned Coverage Expansion (TCE), a framework that decides how source data should be used, either by directly incorporating target-near transitions or by expanding state coverage through target-aligned generation, guided by theoretical analysis. TCE builds on a dual score-based generative model to synthesize target-consistent transitions over an expanded state region. Extensive experiments across diverse cross-domain environments show that TCE consistently outperforms state-of-the-art cross-domain offline RL baselines.
title Bridging Domain Gaps with Target-Aligned Generation for Offline Reinforcement Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2605.13054