AI Paper Atlas

Research map / Probabilistic and scientific modeling / Treatment effects

Good allocations from bad estimates

Sílvia Casacuberta, Moritz Hardt

ICLR 2026 · Probabilistic and scientific modeling › Treatment effects · approach: Meta-learners and partial identification · 5 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
estimate treatment effects coarsely (accuracy sqrt(epsilon)) instead of epsilon before top-K allocationLEA algorithm · five RCT datasets (STAR, TUP, NSW, Acupuncture, Postoperative)allocation value vs samplesImproved
use LEA with coarse estimates (gamma = 0.5, N = M ln(2M/delta)/epsilon)LEA algorithm · five RCT datasets, all budgetsfailure rate of (1-epsilon)-optimal allocation: below 5%Improved
use LEA on small datasets (Acupuncture, Postoperative)LEA algorithm · Acupuncture, Postoperativefailure rateHurt
allow flexible budget K' near K when allocation failsLEA algorithm · five RCT datasetsbudget shift |K-K'|: <= 2 on averageImproved
overspend by adding one extra unit (K+1) on failed budgets (ablation)LEA algorithm · five RCT datasets(1-epsilon)-optimalityImproved

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