AI Paper Atlas

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

Optimal testing using combined test statistics across independent studies

Lasse Vuursteen, Botond Szabo, Aad van der Vaart, Harry van Zanten

NeurIPS 2023 · Learning theory and algorithms › Prediction and decision losses · approach: Adaptive testing and experiments · 2 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
combine local chi-square statistics across m trials instead of using a single trialchi-square combined meta-analysis test · Synthetic many normal means, m=20, n=30, d=2..20ROC (TPR vs FPR)Improved
use pooled data chi-square test instead of combining local chi-square statisticschi-square combined meta-analysis test · Synthetic many normal means, m=20, n=30, d=2..20ROC (TPR vs FPR)Improved
combine local chi-square statistics instead of directional statistics in high dimensionmeta-analysis combination method · Synthetic, d=10 and d=20ROC (TPR vs FPR)Improved
combine directional statistics instead of local chi-square statistics in low dimensionmeta-analysis combination method · Synthetic, d=2ROC (TPR vs FPR)Improved
add coordination between trials via shared randomness (coordinated directional vs uncoordinated directional) (ablation)directional meta-analysis test · Synthetic, d=2 and d=5, small alphaROC (TPR vs FPR)Improved
explore eight additional settings with larger d and mall combination strategies · Synthetic appendix simulationsROC (TPR vs FPR)No clear effect

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