AI Paper Atlas

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

Optimal Decision-Making Based on Prediction Sets

Tao Wang, Edgar Dobriban

ICML 2026 · Learning theory and algorithms › Prediction and decision losses · approach: Algorithms with explicit predictors · 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 ROCP (expected-loss-aware sets and policy) instead of RAC max-minROCP · COVID-19 chest X-ray, 10x severe-mismatch loss matrixworst-case risk and realized lossImproved
use ROCP instead of RACROCP · COVID-19 chest X-ray, 10x severe-mismatch loss matrixcritical mistake rateImproved
use ROCP instead of RACROCP · COVID-19 chest X-ray, baseline loss matrixworst-case riskNo clear effect
use ROCP instead of RACROCP · COVID-19 chest X-ray, baseline loss matrixaverage realized lossHurt
use ROCP instead of RACROCP · COVID-19 chest X-ray, baseline loss matrixcritical mistake rateImproved
use best-response on calibrated model instead of ROCPbest-response baseline · COVID-19 chest X-ray, baseline loss matrixcritical mistake rateHurt
swap in-set max-min rule for the ROCP action rule on LAS/APS/SOCOP sets (ablation)conformal sets with a_ROCP vs a_RA · COVID-19 chest X-ray, baseline loss matrixworst-case risk and realized lossNo clear effect
apply a_ROCP instead of a_RA to LAS/APS/SOCOP sets (ablation)conformal sets · COVID-19 chest X-ray, 10x severe-mismatch loss matrix, alpha >= 0.05worst-case risk and realized lossImproved
use ROCP instead of baselinesROCP · BDD100K toy hazard decisionworst-case risk and realized lossImproved
apply a_ROCP instead of a_RA to conformal sets (ablation)LAS/APS/SOCOP sets · BDD100K toy hazard decision, alpha >= 0.03worst-case certificate and realized lossImproved
use ROCP instead of RACROCP · BDD100K toy hazard decisioncritical mistake rateImproved

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