AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Finite Population Regression Adjustment and Non-asymptotic Guarantees for Treatment Effect Estimation

Mehrdad Ghadiri, David Arbour, Tung Mai, Cameron N Musco, Anup Rao

NeurIPS 2023 · Learning theory and algorithms › Prediction and decision losses · approach: Adaptive testing and experiments · 0 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 cross-adjustment with a random-vector linear regressor instead of uniform HT assignmentATE estimator · BostonATE abs error: 1.736 -> 0.628Improved
use proposed estimator instead of uniform HTATE estimator · IHDPATE abs error: 0.272 -> 0.040Improved
use proposed estimator instead of uniform HTATE estimator · TwinsATE abs error: 1.351e−3 -> 1.218e−3Improved
use proposed estimator versus 4-vectors cross-adjustmentATE estimator · BostonATE abs error: 1.677 -> 0.628Improved
use proposed estimator versus 4-vectors approachATE estimator · IHDPATE abs error: 0.264 -> 0.040Improved
use proposed estimator versus biased leverage score methodATE estimator · IHDPATE abs error: 0.536 -> 0.040Improved
proposed estimator versus GSW designATE estimator · IHDPATE abs error: 0.042 -> 0.040No clear effect
proposed estimator versus GSW on synthetic ATE dataATE estimator · synthetic ATErelative ATE errorHurt
biased classic regression adjustment versus proposed unbiased estimatorATE estimator · BostonATE abs error: 0.333 -> 0.628Hurt
proposed ITE estimator versus leverage score basedITE estimator · IHDPITE error: 2.449 -> 1.765Improved
proposed ITE estimator versus leverage score basedITE estimator · TwinsITE error: 0.156 -> 0.156No clear effect
proposed ITE estimator versus leverage score basedITE estimator · BostonITE error: 0.827 -> 7.654Hurt

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