ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2402.04840
23
0

Efficient Estimation of a Gaussian Mean with Local Differential Privacy

7 February 2024
Nikita Kalinin
Lukas Steinberger
    FedML
ArXivPDFHTML
Abstract

In this paper we study the problem of estimating the unknown mean θ\thetaθ of a unit variance Gaussian distribution in a locally differentially private (LDP) way. In the high-privacy regime (ϵ≤0.67\epsilon\le 0.67ϵ≤0.67), we identify the exact optimal privacy mechanism that minimizes the variance of the estimator asymptotically. It turns out to be the extraordinarily simple sign mechanism that applies randomized response to the sign of Xi−θX_i-\thetaXi​−θ. However, since this optimal mechanism depends on the unknown mean θ\thetaθ, we employ a two-stage LDP parameter estimation procedure which requires splitting agents into two groups. The first n1n_1n1​ observations are used to consistently but not necessarily efficiently estimate the parameter θ\thetaθ by θ~n1\tilde{\theta}_{n_1}θ~n1​​. Then this estimate is updated by applying the sign mechanism with θ~n1\tilde{\theta}_{n_1}θ~n1​​ instead of θ\thetaθ to the remaining n−n1n-n_1n−n1​ observations, to obtain an LDP and efficient estimator of the unknown mean.

View on arXiv
Comments on this paper