Minimax Optimal Procedures for Locally Private Estimation
From MaRDI portal
Statistical aspects of information-theoretic topics (62B10) Minimax procedures in statistical decision theory (62C20) Sampling theory, sample surveys (62D05) Nonparametric estimation (62G05) Density estimation (62G07) Generalized linear models (logistic models) (62J12) Sequential statistical analysis (62L10)
Abstract: Working under a model of privacy in which data remains private even from the statistician, we study the tradeoff between privacy guarantees and the risk of the resulting statistical estimators. We develop private versions of classical information-theoretic bounds, in particular those due to Le Cam, Fano, and Assouad. These inequalities allow for a precise characterization of statistical rates under local privacy constraints and the development of provably (minimax) optimal estimation procedures. We provide a treatment of several canonical families of problems: mean estimation and median estimation, generalized linear models, and nonparametric density estimation. For all of these families, we provide lower and upper bounds that match up to constant factors, and exhibit new (optimal) privacy-preserving mechanisms and computationally efficient estimators that achieve the bounds. Additionally, we present a variety of experimental results for estimation problems involving sensitive data, including salaries, censored blog posts and articles, and drug abuse; these experiments demonstrate the importance of deriving optimal procedures.
Recommendations
- Comment
- Comment
- Comment
- Comment
- Comment
- Optimal Schemes for Discrete Distribution Estimation Under Locally Differential Privacy
- Optimal locally private estimation under \(\ell_p\) loss for \(1\le p\le 2\)
- Optimal Privacy-Aware Estimation
- Empirical risk minimization in the non-interactive local model of differential privacy
- Optimal Local Differentially Private Quantization
Cites work
- A statistical framework for differential privacy
- A unified framework for high-dimensional analysis of M-estimators with decomposable regularizers
- Acceleration of Stochastic Approximation by Averaging
- Bounds on the sample complexity for private learning and private data release
- Convergence of estimates under dimensionality restrictions
- Differential privacy and robust statistics
- Differentially private empirical risk minimization
- Distributed Private Data Analysis: Simultaneously Solving How and What
- Elements of Information Theory
- High-dimensional regression with noisy and missing data: provable guarantees with nonconvexity
- scientific article; zbMATH DE number 3907540 (Why is no real title available?)
- scientific article; zbMATH DE number 44577 (Why is no real title available?)
- scientific article; zbMATH DE number 48436 (Why is no real title available?)
- scientific article; zbMATH DE number 1249686 (Why is no real title available?)
- scientific article; zbMATH DE number 1064667 (Why is no real title available?)
- scientific article; zbMATH DE number 5485574 (Why is no real title available?)
- Inference using noisy degrees: differentially private -model and synthetic graphs
- Information-theoretic determination of minimax rates of convergence
- Interpretations of Directed Information in Portfolio Theory, Data Compression, and Hypothesis Testing
- Introduction to nonparametric estimation
- Nonparametric curve estimation. Methods, theory, and applications
- On optimal and data-based histograms
- On the Fundamental Limits of Adaptive Sensing
- Optimal Rates of Convergence for Deconvolving a Density
- Privacy-preserving statistical estimation with optimal convergence rates
- Randomized response: a survey technique for eliminating evasive answer bias
- Robust Stochastic Approximation Approach to Stochastic Programming
- Statistics for high-dimensional data. Methods, theory and applications.
- Theory of Cryptography
- What can we learn privately?
Cited in
(70)- Providing accurate models across private partitioned data: secure maximum likelihood estimation
- A criterion for privacy protection in data collection and its attainment via randomized response procedures
- On density estimation at a fixed point under local differential privacy
- Strongly universally consistent nonparametric regression and classification with privatised data
- The cost of privacy: optimal rates of convergence for parameter estimation with differential privacy
- On the power of multiple anonymous messages: frequency estimation and selection in the shuffle model of differential privacy
- Minimax optimal goodness-of-fit testing for densities and multinomials under a local differential privacy constraint
- On the universal consistency of histograms anonymised by a randomised response technique
- Obtaining minimax lower bounds: a review
- Statistic selection and MCMC for differentially private Bayesian estimation
- Local differential privacy: elbow effect in optimal density estimation and adaptation over Besov ellipsoids
- An information theoretic approach to post randomization methods under differential privacy
- Geometrizing rates of convergence under local differential privacy constraints
- Principal component analysis in the local differential privacy model
- Tight lower bound of sparse covariance matrix estimation in the local differential privacy model
- Optimal locally private estimation under \(\ell_p\) loss for \(1\le p\le 2\)
- Differentially private precision matrix estimation
- Extremal mechanisms for local differential privacy
- Asymptotically Optimal and Private Statistical Estimation
- Constrained forms of statistical minimax: computation, communication, and privacy
- scientific article; zbMATH DE number 6860834 (Why is no real title available?)
- Comment
- Finite sample differentially private confidence intervals
- Privacy-preserving parametric inference: a case for robust statistics
- Private sequential learning
- Optimal Privacy-Aware Estimation
- Private Sampling: A Noiseless Approach for Generating Differentially Private Synthetic Data
- The structure of optimal private tests for simple hypotheses
- Privacy aware learning
- Topics and Techniques in Distribution Testing: A Biased but Representative Sample
- Multivariate density estimation from privatised data: universal consistency and minimax rates
- Phase transitions for support recovery under local differential privacy
- Minimax randomized response methods for protecting respondent’s privacy
- Distribution-invariant differential privacy
- Goodness-of-fit testing for Hölder continuous densities under local differential privacy
- The right complexity measure in locally private estimation: it is not the Fisher information
- Density estimation under local differential privacy and Hellinger loss
- Interactive versus noninteractive locally differentially private estimation: two elbows for the quadratic functional
- Canonical noise distributions and private hypothesis tests
- On robustness and local differential privacy
- Differentially private inference via noisy optimization
- Private measures, random walks, and synthetic data
- Optimal and differentially private data acquisition: central and local mechanisms
- Gaussian differentially private robust mean estimation and inference
- Distributed optimal subsampling for quantile regression with massive data
- General inferential limits under differential and pufferfish privacy
- Majority vote for distributed differentially private sign selection
- Wasserstein convergence in Bayesian and frequentist deconvolution models
- Nonparametric spectral density estimation under local differential privacy
- Efficiency in local differential privacy
- Fine-filter: an effective defense against poisoning attacks on frequency estimation under LDP
- Enhancements of communication-efficient distributed statistical inference and its privacy preservation
- On the power of private likelihood-ratio tests for goodness-of-fit in frequency tables
- MCMC for Bayesian Nonparametric Mixture Modeling Under Differential Privacy
- General Gaussian noise mechanisms and their optimality for unbiased mean estimation
- Minimax rate for multivariate data under componentwise local differential privacy constraints
- Local differential privacy in survival analysis using private failure indicators
- Rate-Optimal Rank Aggregation with Private Pairwise Rankings
- Differentially private multivariate statistics with an application to contingency table analysis
- Locally private causal inference for randomized experiments
- On rate optimal private regression under local differential privacy
- Faster rates of private stochastic convex optimization
- Evolving privacy: drift parameter estimation for discretely observed i.i.d. diffusion processes under LDP
- Versatile differentially private learning for general loss functions
- Additive-effect assisted learning
- A practical privacy protection procedure for gathering categorical data
- Fundamental limits of membership inference attacks on machine learning models
- Divergence-private query release and high-dimensional mean estimation
- Locally differentially private two-sample testing
- On minimax estimation in the presence of side information about remote data
This page was built for publication: Minimax Optimal Procedures for Locally Private Estimation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4690944)
