On robustness and local differential privacy
From MaRDI portal
(Redirected from On robustness and local differential privacy)
Abstract: It is of soaring demand to develop statistical analysis tools that are robust against contamination as well as preserving individual data owners' privacy. In spite of the fact that both topics host a rich body of literature, to the best of our knowledge, we are the first to systematically study the connections between the optimality under Huber's contamination model and the local differential privacy (LDP) constraints. In this paper, we start with a general minimax lower bound result, which disentangles the costs of being robust against Huber's contamination and preserving LDP. We further study four concrete examples: a two-point testing problem, a potentially-diverging mean estimation problem, a nonparametric density estimation problem and a univariate median estimation problem. For each problem, we demonstrate procedures that are optimal in the presence of both contamination and LDP constraints, comment on the connections with the state-of-the-art methods that are only studied under either contamination or privacy constraints, and unveil the connections between robustness and LDP via partially answering whether LDP procedures are robust and whether robust procedures can be efficiently privatised. Overall, our work showcases a promising prospect of joint study for robustness and local differential privacy.
Recommendations
- Minimax Optimal Procedures for Locally Private Estimation
- Privacy-preserving parametric inference: a case for robust statistics
- Differential privacy and robust statistics
- On density estimation at a fixed point under local differential privacy
- Geometrizing rates of convergence under local differential privacy constraints
Cites work
- A general decision theory for Huber's \(\epsilon\)-contamination model
- A statistical framework for differential privacy
- Challenging the empirical mean and empirical variance: a deviation study
- Combinatorial methods in density estimation
- Differential privacy and robust statistics
- Extremal mechanisms for local differential privacy
- Geometrizing rates of convergence under local differential privacy constraints
- Geometrizing rates of convergence. II
- scientific article; zbMATH DE number 3954047 (Why is no real title available?)
- Interactive Inference Under Information Constraints
- Introduction to nonparametric estimation
- Local differential privacy: elbow effect in optimal density estimation and adaptation over Besov ellipsoids
- Mathematical foundations of infinite-dimensional statistical models
- Minimax optimal goodness-of-fit testing for densities and multinomials under a local differential privacy constraint
- Minimax Optimal Procedures for Locally Private Estimation
- Nonparametric goodness-of-fit testing under Gaussian models
- Private Hypothesis Selection
- Randomized response: a survey technique for eliminating evasive answer bias
- Rates of convergence of minimum distance estimators and Kolmogorov's entropy
- Robust and private Bayesian inference
- Robust confidence limits
- Robust covariance and scatter matrix estimation under Huber's contamination model
- Robust estimators in high-dimensions without the computational intractability
- Robust multivariate mean estimation: the optimality of trimmed mean
- Robust Statistics
- Robust statistics. Theory and methods (with R)
- Strongly universally consistent nonparametric regression and classification with privatised data
- Sub-Gaussian estimators of the mean of a random vector
- Ten Lectures on Wavelets
- The cost of privacy: optimal rates of convergence for parameter estimation with differential privacy
- The structure of optimal private tests for simple hypotheses
- Theory of Cryptography
Cited in
(12)- scientific article; zbMATH DE number 7233838 (Why is no real title available?)
- Heavy Hitters and the Structure of Local Privacy
- Exponential Separations in Local Privacy
- Separating Local & Shuffled Differential Privacy via Histograms
- Local differential privacy protocol for making key-value data robust against poisoning attacks
- On the power of private likelihood-ratio tests for goodness-of-fit in frequency tables
- Minimax rate for multivariate data under componentwise local differential privacy constraints
- Locally robust implementation and its limits
- Review of mathematical optimization in federated learning
- Evolving privacy: drift parameter estimation for discretely observed i.i.d. diffusion processes under LDP
- Privacy guarantees in posterior sampling under contamination
- Locally differentially private two-sample testing
This page was built for publication: On robustness and local differential privacy
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6172196)
