paper-with-me

홈 › Papers

Cross-Domain Policy Adaptation by Capturing Representation Mismatch

2024-05-24 · Jiafei Lyu, Chenjia Bai, Jingwen Yang, Zongqing Lu, Xiu Li

It is vital to learn effective policies that can be transferred to different domains with dynamics discrepancies in reinforcement learning (RL). In this paper, we consider dynamics adaptation settings where there exists dynamics mismatch between the source domain and the target domain, and one can get access to sufficient source domain data, while can only have limited interactions with the target domain. Existing methods address this problem by learning domain classifiers, performing data filtering from a value discrepancy perspective, etc. Instead, we tackle this challenge from a decoupled representation learning perspective. We perform representation learning only in the target domain and measure the representation deviations on the transitions from the source domain, which we show can be a signal of dynamics mismatch. We also show that representation deviation upper bounds performance difference of a given policy in the source domain and target domain, which motivates us to adopt representation deviation as a reward penalty. The produced representations are not involved in either policy or value function, but only serve as a reward penalizer. We conduct extensive experiments on environments with kinematic and morphology mismatch, and the results show that our method exhibits strong performance on many tasks. Our code is publicly available at https://github.com/dmksjfl/PAR.

📄 PDF Abstract BibTeX arXiv:2405.15369

Code (1)

dmksjfl/par 공식 구현 pytorch

Tasks

Reinforcement Learning (RL)Representation Learning

Similar Papers 제목 키워드 기반

In-Context Policy Adaptation via Cross-Domain Skill Diffusion

2025-09-04 · Minjong Yoo, Woo Kyung Kim, Honguk Woo arxiv

In this work, we present an in-context policy adaptation (ICPAD) framework designed for long-horizon multi-task environments, exploring diffusion-based skill learning techniques in cross-domain settings. The framework en…

Reinforcement LearningAutonomous Driving

Domain Adaptation for Reinforcement Learning on the Atari

2018-12-18 · Thomas Carr, Maria Chli, George Vogiatzis

Deep reinforcement learning agents have recently been successful across a variety of discrete and continuous control tasks; however, they can be slow to train and require a large number of interactions with the environme…

continuous-controlContinuous ControlDeep Reinforcement LearningDomain Adaptation+4

Sliced Wasserstein Discrepancy in Disentangling Representation and Adaptation Networks for Unsupervised Domain Adaptation

2025-04-03 · Joel Sol, Shadi Alijani, Homayoun Najjaran

This paper introduces DRANet-SWD, an extension of existing work that disentangles content and style representations of images for unsupervised domain adaptation (UDA). The approach builds upon DRANet by incorporating the…

Domain AdaptationUnsupervised Domain Adaptation

Domain Adaptation In Reinforcement Learning Via Latent Unified State Representation

2021-02-10 · Jinwei Xing, Takashi Nagata, Kexin Chen, Xinyun Zou 외

Despite the recent success of deep reinforcement learning (RL), domain adaptation remains an open problem. Although the generalization ability of RL agents is critical for the real-world applicability of Deep RL, zero-sh…

Autonomous DrivingDeep Reinforcement LearningDomain AdaptationImage-to-Image Translation+5

AdaRL: What, Where, and How to Adapt in Transfer Reinforcement Learning

2021-07-06 · ICLR 2022 4 · Biwei Huang, Fan Feng, Chaochao Lu, Sara Magliacane 외

One practical challenge in reinforcement learning (RL) is how to make quick adaptations when faced with new environments. In this paper, we propose a principled framework for adaptive RL, called \textit{AdaRL}, that adap…

Atari Gamesreinforcement-learningReinforcement Learning (RL)Transfer Reinforcement Learning