A Note on the Prediction-Powered Bootstrap

Tijana Zrnic
5/28/2024
stat.MLcs.LGstat.ME

No official code found. Best implementation: aangelopoulos/ppi_py (301 stars, MIT, last commit 2026-04-10).

Abstract

We introduce PPBoot: a bootstrap-based method for prediction-powered inference. PPBoot is applicable to arbitrary estimation problems and is very simple to implement, essentially only requiring one application of the bootstrap. Through a series of examples, we demonstrate that PPBoot often performs nearly identically to (and sometimes better than) the earlier PPI(++) method based on asymptotic normality$\unicode{x2013}$when the latter is applicable$\unicode{x2013}$without requiring any asymptotic characterizations. Given its versatility, PPBoot could simplify and expand the scope of application of prediction-powered inference to problems where central limit theorems are hard to prove.

View on arXivView PDF

Code Implementations(1)

aangelopoulos/ppi_pyCommunity100%
MIT

A package for statistically rigorous scientific discovery using machine learning. Implements prediction-powered inference.

30140PythonAug 24, 20236 months agoMIT
confidence-intervalconfidence-intervalsinferencemachine-learningp-value+4 more

Similar Papers

PPI++: Efficient Prediction-Powered Inference

84% match1 repo3012023

No Free Lunch: Non-Asymptotic Analysis of Prediction-Powered Inference

83% match2025

Demystifying Prediction Powered Inference

82% match2026

Bayesian Prediction-Powered Inference

81% match1 repo2024

Prediction-Powered Inference with Inverse Probability Weighting

80% match2025

Anytime-valid, Bayes-assisted, Prediction-Powered Inference

80% match2025

Cite this paper

@article{zrnic2024note,
  title  = {A Note on the Prediction-Powered Bootstrap},
  author = {Tijana Zrnic},
  year   = {2024},
  eprint = {2405.18379},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2405.18379v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2405.18379.svg)](https://codeofpaper.com/paper/2405.18379)

Discussion