Universal Private Estimators

11/04/2021
βˆ™
by   Wei Dong, et al.
βˆ™
0
βˆ™

We present universal estimators for the statistical mean, variance, and scale (in particular, the interquartile range) under pure differential privacy. These estimators are universal in the sense that they work on an arbitrary, unknown distribution 𝒫 over ℝ, while yielding strong utility guarantees except for ill-behaved 𝒫. For certain distribution families like Gaussians or heavy-tailed distributions, we show that our universal estimators match or improve existing estimators, which are often specifically designed for the given family and under priori boundedness assumptions on the mean and variance of 𝒫. The removal of these boundedness assumptions is surprising, as existing work believes that they are necessary under pure differential privacy.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
βˆ™ 01/30/2023

A Bias-Variance-Privacy Trilemma for Statistical Estimation

The canonical algorithm for differentially private mean estimation is to...
research
βˆ™ 06/27/2019

Differentially private sub-Gaussian location estimators

We tackle the problem of estimating a location parameter with differenti...
research
βˆ™ 11/30/2022

The Bounded Gaussian Mechanism for Differential Privacy

The Gaussian mechanism is one differential privacy mechanism commonly us...
research
βˆ™ 02/03/2023

From Robustness to Privacy and Back

We study the relationship between two desiderata of algorithms in statis...
research
βˆ™ 05/05/2022

Optimal Algorithms for Mean Estimation under Local Differential Privacy

We study the problem of mean estimation of β„“_2-bounded vectors under the...
research
βˆ™ 06/26/2023

About the Cost of Global Privacy in Density Estimation

We study non-parametric density estimation for densities in Lipschitz an...

Please sign up or login with your details

Forgot password? Click here to reset