AI Paper Atlas

Research map / Reinforcement learning / Imitation learning

LobsDICE: Offline Learning from Observation via Stationary Distribution Correction Estimation

Geon-Hyeong Kim, Jongmin Lee, Youngsoo Jang, Hongseok Yang, Kee-Eung Kim

NeurIPS 2022 · Reinforcement learning › Imitation learning · approach: Offline imitation from imperfect data · 1 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
Match expert state-transition stationary distributions rather than only state stationary distributions (ablation)LobsDICE · Offline LfO state occupancy matching exampleexpert behavior recoveryImproved
Use LobsDICE instead of OPOLOOffline LfO · Random MDPs with stochastic transitionsTV distanceImproved
Use LobsDICE instead of BCOOffline LfO · Random MDPs with stochastic transitionsTV distanceImproved
Use LobsDICE instead of DemoDICEfOOffline LfO · Random MDPs with stochastic transitionsTV distanceImproved
Use LobsDICE instead of OPOLOOffline LfO · Gym-MuJoCo continuous control tasksnormalized scoreImproved
Use LobsDICE instead of IQ-LearnOffline LfO · Gym-MuJoCo continuous control tasksnormalized scoreImproved
Increase the proportion of non-expert data in imperfect demonstrationsBCO and DemoDICEfO · Gym-MuJoCo continuous control tasksnormalized scoreHurt
Increase the amount of dataBCO and DemoDICEfO · Random MDPs with stochastic transitionsTV distanceNo clear effect
Apply state-only behavior cloning to imperfect demonstrations without expert distribution matchingBC on imperfect demonstrations · Gym-MuJoCo continuous control tasksnormalized scoreHurt

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