AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

A Minimax Approach for Optimal Intervention Policy Learning with Two-Stage Outcomes

Chenyang Li, Hao Mei, Yue Liu

ICML 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Confounding-robust policy evaluation

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 principal-strata minimax estimator instead of plain OR upliftOR · ML-1Mpolicy return: 111,510 -> 136,268Improved
use principal-strata minimax estimator instead of plain IPSIPS · ML-1Mpolicy return: 112,620 -> 147,381Improved
use principal-strata minimax estimator instead of plain DRDR · ML-1Mpolicy return: 104,230 -> 150,442Improved
use principal-strata minimax estimator instead of plain OROR · Jesterpolicy return: 102,985 -> 147,923Improved
use principal-strata minimax estimator instead of plain IPSIPS · Jesterpolicy return: 83,555 -> 131,291Improved
use principal-strata minimax estimator instead of plain DRDR · Jesterpolicy return: 99,916 -> 131,465Improved
use uplift (causal) estimation instead of naive outcome-based allocationNAIVE · ML-1Mpolicy return: 42,670 -> 111,510Improved
use uplift (causal) estimation instead of naive outcome-based allocationNAIVE · Jesterpolicy return: 77,310 -> 102,985Improved

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