Environment Agnostic Representation for Visual Reinforcement learning

Hyesong Choi, Hunsang Lee, Seongwon Jeong, Dongbo Min
2/14/2026

No official code found. Best implementation: doihye/EAR (16 stars, last commit 2023-10-18).

Abstract

Generalization capability of vision-based deep reinforcement learning (RL) is indispensable to deal with dynamic environment changes that exist in visual observations. The high-dimensional space of the visual input, however, imposes challenges in adapting an agent to unseen environments. In this work, we propose Environment Agnostic Reinforcement learning (EAR), which is a compact framework for domain generalization of the visual deep RL. Environmentagnostic features (EAFs) are extracted by leveraging three novel objectives based on feature factorization, reconstruction, and episode-aware state shifting, so that policy learning is accomplished only with vital features. EAR is a simple single-stage method with a low model complexity and a fast inference time, ensuring a high reproducibility, while attaining state-of-the-art performance in the DeepMind Control Suite and DrawerWorld benchmarks. Code is available at: https://github.com/doihye/EAR.

DOISemantic Scholar

Code Implementations(1)

doihye/EARCommunity65%

Official pytorch implementation of the paper [Environment Agnostic Representation for Visual Reinforcement learning]

160PythonAug 18, 20233 years ago
domain-generalizationreinforcement-learningvisual-reinforcement-learning

Similar Papers

Salience-Invariant Consistent Policy Learning for Generalization in Visual Reinforcement Learning

85% match2025

Unsupervised Salient Patch Selection for Data-Efficient Reinforcement Learning

84% match2024

A Simple Framework for Generalization in Visual RL under Dynamic Scene Perturbations

84% match2026

Self-Consistent Model-based Adaptation for Visual Reinforcement Learning

83% match2025

Vision-Language Models Provide Promptable Representations for Reinforcement Learning

83% match2024

Learning to Identify Critical States for Reinforcement Learning from Videos

82% match1 repo282026

Cite this paper

@article{choi2026environment,
  title  = {Environment Agnostic Representation for Visual Reinforcement learning},
  author = {Hyesong Choi and Hunsang Lee and Seongwon Jeong and Dongbo Min},
  year   = {2026},
  doi    = {10.1109/ICCV51070.2023.00031},
  url    = {https://doi.org/10.1109/ICCV51070.2023.00031},
  journal = {ICCV 2023 2023}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/0de41687f049cc3ae310ec5119ac0634070dffa4.svg)](https://codeofpaper.com/paper/0de41687f049cc3ae310ec5119ac0634070dffa4)

Discussion