AI Paper Atlas

Research map / Reinforcement learning / Imitation learning

An Empirical Investigation of Representation Learning for Imitation

Xin Chen, Sam Toyer, Cody Wild, Scott Emmons, Ian Fischer, Kuang-Huei Lee, Neel Alex, Steven H Wang, Ping Luo, Stuart Russell, Pieter Abbeel, Rohin Shah

NeurIPS 2021 · Reinforcement learning › Imitation learning · approach: Offline imitation from imperfect data · 6 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 image augmentations to BC (ablation)Behavioral Cloning · DMC finger-spinreturn: 940±4* -> 730±9Hurt
Add image augmentations to BC (ablation)Behavioral Cloning · DMC reacher-easyreturn: 452±34 -> 874±21Improved
Add image augmentations to BC (ablation)Behavioral Cloning · Procgen fruitbot-trainreturn: 11.4±0.6 -> 18.3±1.9Improved
Add image augmentations to BC (ablation)Behavioral Cloning · MAGICAL MatchRegionsscore: 0.28±0.08 -> 0.43±0.02Improved
Use representation learning pretraining instead of augmentation-only BCBC across the evaluated benchmarks · DMC, Procgen, and MAGICALreturnMixed
Add joint representation learning as an auxiliary lossBC on DMC · DMCreturnImproved
Add representation learning pretrainingBC on MAGICAL MoveToRegion · MAGICAL MoveToRegionscore: 0.74 -> 0.83Improved
Remove discriminator augmentations (ablation)GAIL on DMC finger-spin · DMC finger-spinreturn: 868±12 -> 0±0Hurt

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