Robustness Implies Privacy in Statistical Estimation
From MaRDI portal
Abstract: We study the relationship between adversarial robustness and differential privacy in high-dimensional algorithmic statistics. We give the first black-box reduction from privacy to robustness which can produce private estimators with optimal tradeoffs among sample complexity, accuracy, and privacy for a wide range of fundamental high-dimensional parameter estimation problems, including mean and covariance estimation. We show that this reduction can be implemented in polynomial time in some important special cases. In particular, using nearly-optimal polynomial-time robust estimators for the mean and covariance of high-dimensional Gaussians which are based on the Sum-of-Squares method, we design the first polynomial-time private estimators for these problems with nearly-optimal samples-accuracy-privacy tradeoffs. Our algorithms are also robust to a constant fraction of adversarially-corrupted samples.
Cited in
(4)- General inferential limits under differential and pufferfish privacy
- Mixtures of Gaussians are privately learnable with a polynomial number of samples
- Differentially private learning beyond the classical dimensionality regime
- The full landscape of robust mean testing: sharp separations between oblivious and adaptive contamination
This page was built for publication: Robustness Implies Privacy in Statistical Estimation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6420075)
