AI Paper Atlas

Research map / Reinforcement learning / Offline RL

Provably Efficient Causal Reinforcement Learning with Confounded Observational Data

Lingxiao Wang, Zhuoran Yang, Zhaoran Wang

NeurIPS 2021 · Reinforcement learning › Offline RL · approach: Diffusion models for offline decision making · 14 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 backdoor-adjusted observational data to online value iterationDOVI · Confounded MDP with partially observed confoundersregret: When the observational data are sufficiently informative upon the adjustments, \Delta_H decreases towards zeroImproved
Add frontdoor-adjusted observational data through an observed intermediate stateDOVI+ · Confounded MDP with unobserved confounders and observed intermediate statesregret: When the observational data are sufficiently informative upon the adjustments, \Delta_{1, H} and \Delta_{2, H} are smallImproved
Use observational data that are unavailable or uninformative after adjustmentDOVI · Confounded MDP with partially observed confoundersregret: \Delta_H is a logarithmic factorNo clear effect
Increase the effective sample size of sufficiently informative adjusted observational dataDOVI · Confounded MDP with partially observed confoundersregret: \Delta_H = \cO(\sqrt{\log(1 + K/(n + d))}) = \cO(\sqrt{K/(n + d)})Improved

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