On Privately Estimating a Single Parameter

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Asi, Hilal, Duchi, John C., Talwar, Kunal
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915208289583104
author Asi, Hilal
Duchi, John C.
Talwar, Kunal
author_facet Asi, Hilal
Duchi, John C.
Talwar, Kunal
contents We investigate differentially private estimators for individual parameters within larger parametric models. While generic private estimators exist, the estimators we provide repose on new local notions of estimand stability, and these notions allow procedures that provide private certificates of their own stability. By leveraging these private certificates, we provide computationally and statistical efficient mechanisms that release private statistics that are, at least asymptotically in the sample size, essentially unimprovable: they achieve instance optimal bounds. Additionally, we investigate the practicality of the algorithms both in simulated data and in real-world data from the American Community Survey and US Census, highlighting scenarios in which the new procedures are successful and identifying areas for future work.
format Preprint
id arxiv_https___arxiv_org_abs_2503_17252
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On Privately Estimating a Single Parameter
Asi, Hilal
Duchi, John C.
Talwar, Kunal
Machine Learning
Cryptography and Security
Statistics Theory
We investigate differentially private estimators for individual parameters within larger parametric models. While generic private estimators exist, the estimators we provide repose on new local notions of estimand stability, and these notions allow procedures that provide private certificates of their own stability. By leveraging these private certificates, we provide computationally and statistical efficient mechanisms that release private statistics that are, at least asymptotically in the sample size, essentially unimprovable: they achieve instance optimal bounds. Additionally, we investigate the practicality of the algorithms both in simulated data and in real-world data from the American Community Survey and US Census, highlighting scenarios in which the new procedures are successful and identifying areas for future work.
title On Privately Estimating a Single Parameter
topic Machine Learning
Cryptography and Security
Statistics Theory
url https://arxiv.org/abs/2503.17252