.png){kind=icon}
.png){kind=icon}
Return
Prediction-Powered Model Checking via Predictiveness Comparisons
.png){kind=icon}
PRE
AI
DOI：10.1007/s11424-026-5075-1.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Model checking evaluates whether a statistical model faithfully captures the underlying data-generating process. Classical tests — such as local-smoothing and empirical-process methods — break down in high dimensions. More recent approaches use predictiveness comparisons with flexible machine-learning model fitting procedures to yield algorithm-agnostic tests, yet they require large labeled samples. The authors introduce a prediction-powered, semi-supervised framework that: 1) Imputes responses for unlabeled data via a pretrained model; 2) Corrects imputation bias with a rectifier calibrated on labeled data; 3) Adaptively balances these components through a data-driven power-tuning parameter. Building on algorithm-agnostic out-of-sample predictiveness comparisons, the proposed method integrates unlabeled information to enhance power. Theoretical analyses and numerical results demonstrate that the proposed test controls Type I error and substantially improves power over fully supervised counterparts, even under imputation-model misspecification.
Keywords:
Algorithm-agnostic inference
asymptotic normality
model checking
prediction-powered inference
semi-supervised inference
Journal
IF:
2.8
Papers:
212
Citations:
2.1K