EUVIMEDEUVIMED entdecken
Crossref

PPI++: Efficient prediction-powered inference

Anastasios N. Angelopoulos, John C. Duchi, Tijana Zrnic

The Annals of Applied Statistics · 01.09.2026 · 20 (3) · 2235-2252

DOI: 10.1214/26-aoas2215 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Annals of Applied Statistics, erschienen am 01.09.2026. Verfasst von Anastasios N. Angelopoulos und 2 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 38 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present PPI++: a computationally lightweight methodology for estimation and inference based on a small labeled dataset and a typically much larger dataset of machine-learning predictions. The methods automatically adapt to the quality of available predictions, yielding easy-to-compute confidence sets—for parameters of any dimensionality—that always improve on classical intervals using only the labeled data. PPI++ builds on prediction-powered inference (PPI), which targets the same problem setting, improving its computational and statistical efficiency. Real and synthetic experiments demonstrate the benefits of the proposed adaptations.

Zitieren

APA

Angelopoulos, A. N., Duchi, J. C., & Zrnic, T. (2026). PPI++: Efficient prediction-powered inference. The Annals of Applied Statistics, 20(3), 2235-2252. https://doi.org/10.1214/26-aoas2215

Vancouver

Angelopoulos AN, Duchi JC, Zrnic T. PPI++: Efficient prediction-powered inference. The Annals of Applied Statistics. 2026;20(3):2235-2252. doi:10.1214/26-aoas2215

BibTeX
@article{Angelopoulos2026,
  author = {Angelopoulos, Anastasios N. and Duchi, John C. and Zrnic, Tijana},
  title = {PPI++: Efficient prediction-powered inference},
  journal = {The Annals of Applied Statistics},
  year = {2026},
  volume = {20},
  number = {3},
  pages = {2235--2252},
  doi = {10.1214/26-aoas2215}
}
Kostenlos starten →