English

Tight and Robust Private Mean Estimation with Few Users

Data Structures and Algorithms 2022-06-14 v2 Cryptography and Security Machine Learning Statistics Theory Statistics Theory

Abstract

In this work, we study high-dimensional mean estimation under user-level differential privacy, and design an (ε,δ)(\varepsilon,\delta)-differentially private mechanism using as few users as possible. In particular, we provide a nearly optimal trade-off between the number of users and the number of samples per user required for private mean estimation, even when the number of users is as low as O(1εlog1δ)O(\frac{1}{\varepsilon}\log\frac{1}{\delta}). Interestingly, this bound on the number of \emph{users} is independent of the dimension (though the number of \emph{samples per user} is allowed to depend polynomially on the dimension), unlike the previous work that requires the number of users to depend polynomially on the dimension. This resolves a problem first proposed by Amin et al. Moreover, our mechanism is robust against corruptions in up to 49%49\% of the users. Finally, our results also apply to optimal algorithms for privately learning discrete distributions with few users, answering a question of Liu et al., and a broader range of problems such as stochastic convex optimization and a variant of stochastic gradient descent via a reduction to differentially private mean estimation.

Keywords

Cite

@article{arxiv.2110.11876,
  title  = {Tight and Robust Private Mean Estimation with Few Users},
  author = {Hossein Esfandiari and Vahab Mirrokni and Shyam Narayanan},
  journal= {arXiv preprint arXiv:2110.11876},
  year   = {2022}
}

Comments

41 pages. To appear in the International Conference on Machine Learning (ICML), 2022

R2 v1 2026-06-24T07:06:38.180Z