Self-Consistent Model-based Adaptation for Visual Reinforcement Learning

2025-02-14 · Xinning Zhou, Chengyang Ying, Yao Feng, Hang Su, Jun Zhu

Visual reinforcement learning agents typically face serious performance declines in real-world applications caused by visual distractions. Existing methods rely on fine-tuning the policy's representations with hand-crafted augmentations. In this work, we propose Self-Consistent Model-based Adaptation (SCMA), a novel method that fosters robust adaptation without modifying the policy. By transferring cluttered observations to clean ones with a denoising model, SCMA can mitigate distractions for various policies as a plug-and-play enhancement. To optimize the denoising model in an unsupervised manner, we derive an unsupervised distribution matching objective with a theoretical analysis of its optimality. We further present a practical algorithm to optimize the objective by estimating the distribution of clean observations with a pre-trained world model. Extensive experiments on multiple visual generalization benchmarks and real robot data demonstrate that SCMA effectively boosts performance across various distractions and exhibits better sample efficiency.

📄 PDF Abstract BibTeX arXiv:2502.09923

Code (0)

등록된 구현이 없습니다.

Tasks

Denoising reinforcement-learning Reinforcement Learning

Lyapunov-Guided Self-Alignment: Test-Time Adaptation for Offline Safe Reinforcement Learning

2026-04-29 · Seungyub Han, Hyungjin Kim, Jungwoo Lee arxiv

Offline reinforcement learning (RL) agents often fail when deployed, as the gap between training datasets and real environments leads to unsafe behavior. To address this, we present SAS (Self-Alignment for Safety), a tra…

Reinforcement Learning Test-time Adaptation Bayesian Inference

Domain Adaptation with Adaptive Imagination for Visual Reinforcement Learning under Limited Target Data

2026-06-29 · Hyunwoo Park, Sang-Hyun Lee arxiv

Sim-to-real transfer remains a major obstacle for reinforcement learning (RL), especially for vision-based control where image observations exacerbate the state-distribution shift between simulation and the real world. D…

Reinforcement Learning Domain Adaptation

Self-Supervised Sim-to-Real Adaptation for Visual Robotic Manipulation

2019-10-21 · Rae Jeong, Yusuf Aytar, David Khosid, Yuxiang Zhou 외

Collecting and automatically obtaining reward signals from real robotic visual data for the purposes of training reinforcement learning algorithms can be quite challenging and time-consuming. Methods for utilizing unlabe…

Deep Reinforcement Learning Domain Adaptation reinforcement-learning Reinforcement Learning+2

Self-Supervised Visual Prompting for Cross-Domain Road Damage Detection

2025-11-16 · Xi Xiao, Zhuxuanzi Wang, Mingqiao Mo, Chen Liu 외 arxiv

The deployment of automated pavement defect detection is often hindered by poor cross-domain generalization. Supervised detectors achieve strong in-domain accuracy but require costly re-annotation for new environments, w…

Domain Generalization Road Damage Detection

GrndCtrl: Grounding World Models via Self-Supervised Reward Alignment

2025-12-01 · Haoyang He, Jay Patrikar, Dong-Ki Kim, Max Smith 외 arxiv

Recent advances in video world modeling have enabled large-scale generative models to simulate embodied environments with high visual fidelity, providing strong priors for prediction, planning, and control. Yet, despite …

Reinforcement Learning