AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Environment Agnostic Representation for Visual Reinforcement Learning

Hyesong Choi, Hunsang Lee, Seongwon Jeong, Dongbo Min

ICCV 2023 · Reinforcement learning › Visual and representation RL · approach: Robust visual encoders for RL · 11 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add feature separation loss L_sep to SAC (ablation)SAC · DMControl randomized colors, Walker walkepisode return: 412 -> 799Improved
add all three objectives instead of L_sep + L_recon only (ablation)EAR · DMControl randomized colors, Walker walkepisode return: 891 -> 922Improved
add episode-aware state shift loss L_ss (ablation)EAR · DMControl randomized colorsepisode return std: Walker walk std 134 -> 37Improved
add reconstruction loss L_recon alongside L_sep (ablation)EAR · DMControl randomized colorsepisode returnNo clear effect
apply ESS constraint to factorized EAF instead of non-factorized feature (ablation)EAR · DMControl randomized colorsepisode return: Cartpole swingup 805 -> 884; Cheetah run 389 -> 368Mixed
use environment-agnostic feature learning instead of test-time adaptation (PAD)EAR · DMControl randomized colors, Walker walkepisode return: 468 -> 922Improved
EAR versus VAI on video backgroundsEAR · DMControl video backgroundsepisode return: improves 7 of 8 tasks; Finger spin 764 vs 717Mixed
EAR on DrawerClose texture environmentsVAI · DrawerWorld DrawerClose, Fabricsuccess rate: 74 -> 90Improved
EAR versus VAI on DrawerOpen FabricVAI · DrawerWorld DrawerOpen, Fabricsuccess rate: 99 -> 93Hurt
add EAR objectives to SAC encoderSAC · Model complexityparameters (M): 2.54 -> 2.55No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy