Efficient Estimation of a Gaussian Mean with Local Differential Privacy
Abstract
In this paper we study the problem of estimating the unknown mean $\theta$ of a unit variance Gaussian distribution in a locally differentially private (LDP) way. In the highprivacy regime ($\epsilon\le 0.67$), 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 $X_i\theta$. However, since this optimal mechanism depends on the unknown mean $\theta$, we employ a twostage LDP parameter estimation procedure which requires splitting agents into two groups. The first $n_1$ observations are used to consistently but not necessarily efficiently estimate the parameter $\theta$ by $\tilde{\theta}_{n_1}$. Then this estimate is updated by applying the sign mechanism with $\tilde{\theta}_{n_1}$ instead of $\theta$ to the remaining $nn_1$ observations, to obtain an LDP and efficient estimator of the unknown mean.
 Publication:

arXiv eprints
 Pub Date:
 February 2024
 DOI:
 10.48550/arXiv.2402.04840
 arXiv:
 arXiv:2402.04840
 Bibcode:
 2024arXiv240204840K
 Keywords:

 Mathematics  Statistics Theory