AI Paper Atlas

Research map / Probabilistic and scientific modeling / Treatment effects

Optimal Treatment Regimes for Proximal Causal Learning

Tao Shen, Yifan Cui

NeurIPS 2023 · Probabilistic and scientific modeling › Treatment effects · approach: Latent-variable causal inference · 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
use proximal-causal-inference ITRs (using proxies Z and W) instead of causal forest or outcome weighted learningd_z, d_w, d_z-union-w, d_zw · Simulation scenarios 1-6 (n=1000)empirical value functionImproved
combine treatment and outcome confounding bridges with a covariate-dependent mixing rule instead of using only one bridged_zw^pi-hat vs d_z, d_w · Simulation scenarios 1-6empirical value functionImproved
proposed mixed regime versus the union rule d_z-union-wd_zw^pi-hat vs d_zUw · Simulation scenarios 1-6empirical value functionImproved
proposed mixed regime versus single-bridge regimes in a scenario where d_z and d_w mostly agreed_zw^pi-hat · Simulation scenario 5empirical value functionNo clear effect
proposed regime versus existing proximal regimes on real RHC/SUPPORT data, 4 to 10 proxiesd_zw^pi-hat · SUPPORT RHC, 4 settingsestimated value functionImproved
changed sample size and modified behavior policy in simulationsd_zw^pi-hat · Appendix robustness simulationsempirical value functionImproved

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