Back to search

University of Waterloo

A Bias-Variance-Privacy Trilemma for Statistical Estimation

Abstract

dc:description.abstract

The canonical algorithm for differentially private mean estimation is to first clip the samples to a bounded range and then add noise to their empirical mean. Clipping controls the sensitivity and, hence, the variance of the noise that we add for privacy. But clipping also introduces statistical bias. We prove that this tradeoff is inherent: no algorithm can simultaneously have low bias, low variance, and low privacy loss for arbitrary distributions. On the positive side, we show that unbiased mean estimation is possible under approx- imate differential privacy if we assume that the distribution is symmetric. Relaxing to approximate differential privacy is necessary. We show that, even when the data is sampled from a Gaussian, unbiased mean estimation is impossible under pure or concentrated differential privacy.

Degree

thesis:*
Grantor dc:publisher
University of Waterloo
Year dc:date.issued
2023

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Regehr, Matthew
Advisors dc:contributor.advisor
  • Kamath, Gautam
  • Ben-David, Shai

Subjects

dc:subject × 3

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*
Handle dc:identifier.uri
http://hdl.handle.net/10012/19784

Chain of custody

source
Harvested from
University of Waterloo
Base URL
dspacemainprd01.lib.uwaterloo.ca/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Regehr, Matthew. A Bias-Variance-Privacy Trilemma for Statistical Estimation. University of Waterloo, 2023. http://hdl.handle.net/10012/19784