Efficient, Differentially Private Point Estimators

2008/09/27 by Adam Smith, Smith, Adam · 6 citations
Computer Science · Engineering · #Cryptography and Data Security #Privacy-Preserving Technologies in Data #Wireless Communication Security Techniques #cs.CR #cs.DS

paper · pdf · doi:10.48550/arxiv.0809.4794

9 pages

arxiv created 2008/09/27 · arxiv updated 2009/12/01

Abstract

Differential privacy is a recent notion of privacy for statistical databases that provides rigorous, meaningful confidentiality guarantees, even in the presence of an attacker with access to arbitrary side information. We show that for a large class of parametric probability models, one can construct a differentially private estimator whose distribution converges to that of the maximum likelihood estimator. In particular, it is efficient and asymptotically unbiased. This result provides (further) compelling evidence that rigorous notions of privacy in statistical databases can be consistent with statistically valid inference.

Citations

• Algebraic algorithms for sampling from conditional distributions
• Bias reduction of maximum likelihood estimates

Cited by

• Differentially Private Nonparametric Confidence Intervals Under Minimal Distributional Assumptions
• Differentially Private Combinatorial Optimization
• Differential Privacy and Machine Learning: a Survey and Review
• Privacy for Free: Posterior Sampling and Stochastic Gradient Monte Carlo
• Differentially Private Learning of Undirected Graphical Models using\n Collective Graphical Models
• Differentially Private Bayesian Inference for Generalized Linear Models

Related

• Convergence Rates for Differentially Private Statistical Estimation
• Statistical Inference in the Differential Privacy Model
• A statistical framework for differential privacy
• A Bias-Accuracy-Privacy Trilemma for Statistical Estimation
• Perturbed M-Estimation: A Further Investigation of Robust Statistics for Differential Privacy