Translator Disclaimer
October 2020 Geometrizing rates of convergence under local differential privacy constraints
Angelika Rohde, Lukas Steinberger
Ann. Statist. 48(5): 2646-2670 (October 2020). DOI: 10.1214/19-AOS1901

Abstract

We study the problem of estimating a functional $\theta ({\mathbb{P}})$ of an unknown probability distribution ${\mathbb{P}}\in {\mathcal{P}}$ in which the original iid sample $X_{1},\dots ,X_{n}$ is kept private even from the statistician via an $\alpha $-local differential privacy constraint. Let $\omega _{\mathrm{TV}}$ denote the modulus of continuity of the functional $\theta $ over ${\mathcal{P}}$ with respect to total variation distance. For a large class of loss functions $l$ and a fixed privacy level $\alpha $, we prove that the privatized minimax risk is equivalent to $l(\omega _{\mathrm{TV}}(n^{-1/2}))$ to within constants, under regularity conditions that are satisfied, in particular, if $\theta $ is linear and ${\mathcal{P}}$ is convex. Our results complement the theory developed by Donoho and Liu (1991) with the nowadays highly relevant case of privatized data. Somewhat surprisingly, the difficulty of the estimation problem in the private case is characterized by $\omega _{\mathrm{TV}}$, whereas, it is characterized by the Hellinger modulus of continuity if the original data $X_{1},\dots ,X_{n}$ are available. We also find that for locally private estimation of linear functionals over a convex model a simple sample mean estimator, based on independently and binary privatized observations, always achieves the minimax rate. We further provide a general recipe for choosing the functional parameter in the optimal binary privatization mechanisms and illustrate the general theory in numerous examples. Our theory allows us to quantify the price to be paid for local differential privacy in a large class of estimation problems. This price appears to be highly problem specific.

Citation

Download Citation

Angelika Rohde. Lukas Steinberger. "Geometrizing rates of convergence under local differential privacy constraints." Ann. Statist. 48 (5) 2646 - 2670, October 2020. https://doi.org/10.1214/19-AOS1901

Information

Received: 1 May 2019; Revised: 1 August 2019; Published: October 2020
First available in Project Euclid: 19 September 2020

MathSciNet: MR4152116
Digital Object Identifier: 10.1214/19-AOS1901

Subjects:
Primary: 62G05
Secondary: 62C20

Rights: Copyright © 2020 Institute of Mathematical Statistics

JOURNAL ARTICLE
25 PAGES

This article is only available to subscribers.
It is not available for individual sale.
+ SAVE TO MY LIBRARY

SHARE
Vol.48 • No. 5 • October 2020
Back to Top