CoinPress: Practical Private Mean and Covariance Estimation

06/11/2020
by   Sourav Biswas, et al.
0

We present simple differentially private estimators for the mean and covariance of multivariate sub-Gaussian data that are accurate at small sample sizes. We demonstrate the effectiveness of our algorithms both theoretically and empirically using synthetic and real-world datasets—showing that their asymptotic error rates match the state-of-the-art theoretical bounds, and that they concretely outperform all previous methods. Specifically, previous estimators either have weak empirical accuracy at small sample sizes, perform poorly for multivariate data, or require the user to provide strong a priori estimates for the parameters.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/24/2021

Covariance-Aware Private Mean Estimation Without Private Covariance Estimation

We present two sample-efficient differentially private mean estimators f...
research
11/08/2021

A Private and Computationally-Efficient Estimator for Unbounded Gaussians

We give the first polynomial-time, polynomial-sample, differentially pri...
research
02/19/2020

Propose, Test, Release: Differentially private estimation with high probability

We derive concentration inequalities for differentially private median a...
research
06/14/2023

PLAN: Variance-Aware Private Mean Estimation

Differentially private mean estimation is an important building block in...
research
12/05/2019

Context Aware Password Guessability via Multi-Dimensional Rank Estimation

Password strength estimators are used to help users avoid picking weak p...
research
10/12/2022

Concentration of the exponential mechanism and differentially private multivariate medians

We prove concentration inequalities for the output of the exponential me...
research
01/17/2023

A Fast Algorithm for Adaptive Private Mean Estimation

We design an (ε, δ)-differentially private algorithm to estimate the mea...

Please sign up or login with your details

Forgot password? Click here to reset