AI Paper Atlas

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

Tree-Values: Selective Inference for Regression Trees

Anna C. Neufeld, Lucy L. Gao, Daniela M. Witten

JMLR 2022 · Learning theory and algorithms › Prediction and decision losses · approach: Adaptive testing and experiments · 17 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 selective Z-intervals instead of naive Z-intervalsCART tree inference · Simulation, level 2 sibling differencecoverage of 95% CI: 0.410 -> 0.951Improved
use selective Z-intervals instead of naive Z-intervalsCART tree inference · Simulation, level 2 region meancoverage of 95% CI: 0.645 -> 0.950Improved
use selective Z-intervals instead of sample splittingCART tree inference · Simulation, level 1 region meancoverage of 95% CI: 0.918 -> 0.951Improved
use selective Z-test instead of naive Z-test for p-valuesCART tree inference · Global null a=b=0p-value uniformityImproved
estimate sigma with conservative estimator instead of known sigma (ablation)Selective Z-intervals · Simulation, level 1 region meancoverage of 95% CI: 0.95 -> 0.98Hurt
estimate sigma with SSE estimator instead of known sigma (ablation)Selective Z-intervals · Simulation, level 2 region meancoverage of 95% CI: 0.95 -> 0.94Hurt
use conservative sigma estimator for selective Z-test (ablation)Selective Z-test · Simulation power, large bpowerHurt
use SSE sigma estimator instead of known sigma (ablation)Selective Z-test · Simulation powerpowerNo clear effect
condition on ancestors of the region (extra information) (ablation)Selective Z-test for regions · Simulation (appendix)powerNo clear effect
apply selective inference to CART instead of CTree p-valueBox Lunch Study first split hunger<1.8 · Box Lunch Studyp-value: less than 0.001 -> 0.44Mixed

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