AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

CLUTR: Curriculum Learning via Unsupervised Task Representation Learning

Abdus Salam Azad, Izzeddin Gur, Jasper Emhoff, Nathaniel Alexis, Aleksandra Faust, Pieter Abbeel, Ion Stoica

ICML 2023 · Reinforcement learning › Visual and representation RL · approach: Learned state and action representations · 3 citations (OpenAlex)

Read the paper

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
Use CLUTR with flexible regret instead of PAIRED with flexible regretPAIRED on CarRacing · F1 benchmarkreturn: 82%Improved
Use CLUTR with standard regret instead of PAIRED with standard regretPAIRED on CarRacing · F1 benchmarkreturn: 10.6XImproved
Use CLUTR with flexible regret instead of PAIRED with flexible regretPAIRED on MiniGrid · 16 unseen navigation taskssolve rate: 35%Improved
Use CLUTR with standard regret instead of PAIRED with standard regretPAIRED on MiniGrid · 16 unseen navigation taskssolve rate: 45%Improved
Keep the pretrained task decoder fixed during curriculum learning (ablation)CLUTR on CarRacing · CarRacingreturn: 29% performance dropImproved
Keep the pretrained task decoder fixed during curriculum learning (ablation)CLUTR with flexible regret · CarRacingreturn: 10%Improved
Sort task sequences when training the VAE (ablation)CLUTR on CarRacing · CarRacingreturn: 31% performance dropImproved
Use flexible regret instead of standard regretPAIRED on CarRacing · F1 benchmarkreturn: 257 ± 16 vs 26 ± 19Improved
Use flexible regret instead of standard regretCLUTR on CarRacing · F1 benchmarkreturn: 468 ± 21 vs 276 ± 24Improved

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