AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Learning Domain Invariant Representations in Goal-conditioned Block MDPs

Beining Han, Chongyi Zheng, Harris Chan, Keiran Paster, Michael R. Zhang, Jimmy Ba

NeurIPS 2021 · Reinforcement learning › Visual and representation RL · approach: Graph search and compositional goals · 0 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 perfect alignment losses (MMD and difference) with aligned sampling to Skew-FitSkew-Fit · Reach test environmentshand distance: 0.111 -> 0.076Improved
Add perfect alignment to Skew-FitSkew-Fit · Door test environmentsangle difference: 0.194 -> 0.106Improved
Add perfect alignment to Skew-FitSkew-Fit · Push test environmentspuck distance: 0.086 -> 0.069Improved
Add perfect alignment to Skew-FitSkew-Fit · Pickup test environmentsobject distance: 0.037 -> 0.028Improved
Add RAD data augmentation to Skew-FitSkew-Fit · Pickup test environmentsobject distance: 0.037 -> 0.040Hurt
Add RAD data augmentation to Skew-FitSkew-Fit · Door test environmentsangle difference: 0.194 -> 0.162Improved
Use bisimulation methods (MISA, DBC) instead of PA-SFSkew-Fit · Door test environmentsangle difference: 0.106 -> 0.255Hurt
Train PA-SF on training environments vs oracle single-environment Skew-FitPA-SF · training environmentsdistance metrics: 0.058 -> 0.057No clear effect
Remove difference loss L_DIFF (PA-SF w/o D) (ablation)PA-SF · Doorlatent error rate and test performanceHurt
Remove both MMD and difference losses (PA-SF w/o MD) (ablation)PA-SF · Doorlatent error rateHurt
Remove aligned sampling (PA-SF w/o AS) (ablation)PA-SF · Door training environmentstraining performanceHurt
Add MMD loss for alignment (ablation)PA-SF · Doorlatent error rate on training environmentsImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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