General-Purpose Differentially-Private Confidence Intervals

One of the most common statistical goals is to estimate a population parameter and quantify uncertainty by constructing a confidence interval. However, the field of differential privacy lacks easy-to-use and general methods for doing so. We partially fill this gap by developing two broadly applicable methods for private confidence-interval construction. The first is based on asymptotics: for two widely used model classes, exponential families and linear regression, a simple private estimator has the same asymptotic normal distribution as the corresponding non-private estimator, so confidence intervals can be constructed using quantiles of the normal distribution. These are computationally cheap and accurate for large data sets, but do not have good coverage for small data sets. The second approach is based on the parametric bootstrap. It applies "out of the box" to a wide class of private estimators and has good coverage at small sample sizes, but with increased computational cost. Both methods are based on post-processing the private estimator and do not consume additional privacy budget.

[1]  Frank McSherry,et al.  Probabilistic Inference and Differential Privacy , 2010, NIPS.

[2]  Sofya Raskhodnikova,et al.  Smooth sensitivity and sampling in private data analysis , 2007, STOC '07.

[3]  James Honaker,et al.  Bootstrap Inference and Differential Privacy: Standard Errors for Free∗ , 2018 .

[4]  Martin J. Wainwright,et al.  Local privacy and statistical minimax rates , 2013, 2013 51st Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[5]  Or Sheffet,et al.  Differentially Private Ordinary Least Squares , 2015, ICML.

[6]  M. Kenward,et al.  An Introduction to the Bootstrap , 2007 .

[7]  Christos Dimitrakakis,et al.  Robust and Private Bayesian Inference , 2013, ALT.

[8]  David Hinkley,et al.  Bootstrap Methods: Another Look at the Jackknife , 2008 .

[9]  Vito D'Orazio,et al.  Differential Privacy for Social Science Inference , 2015 .

[10]  Stephen T. Joy.  The Differential Privacy of Bayesian Inference , 2015 .

[11]  Larry Wasserman,et al.  All of Statistics: A Concise Course in Statistical Inference , 2004 .

[12]  Ryan M. Rogers,et al.  Differentially Private Chi-Squared Hypothesis Testing: Goodness of Fit and Independence Testing , 2016, ICML 2016.

[13]  Yu-Xiang Wang,et al.  Revisiting differentially private linear regression: optimal and adaptive prediction & estimation in unbounded domain , 2018, UAI.

[14]  Alexander J. Smola,et al.  Privacy for Free: Posterior Sampling and Stochastic Gradient Monte Carlo , 2015, ICML.

[15]  Eftychia Solea,et al.  Differentially Private Hypothesis Testing For Normal Random Variables. , 2014 .

[16]  Cynthia Dwork,et al.  Calibrating Noise to Sensitivity in Private Data Analysis , 2006, TCC.

[17]  Antti Honkela,et al.  Differentially private Bayesian learning on distributed data , 2017, NIPS.

[18]  L. Wasserman,et al.  A Statistical Framework for Differential Privacy , 2008, 0811.2501.

[19]  Aaron Roth,et al.  The Algorithmic Foundations of Differential Privacy , 2014, Found. Trends Theor. Comput. Sci..

[20]  S. T. Buckland,et al.  An Introduction to the Bootstrap. , 1994 .

[21]  Ashwin Machanavajjhala,et al.  Differentially Private Significance Tests for Regression Coefficients , 2017, Journal of Computational and Graphical Statistics.

[22]  Antti Honkela,et al.  Efficient differentially private learning improves drug sensitivity prediction , 2016, Biology Direct.

[23]  Marco Gaboardi,et al.  Locally Private Mean Estimation: Z-test and Tight Confidence Intervals , 2018, AISTATS.

[24]  Andrew T. Levin,et al.  Inferences from Parametric and Non-Parametric Covariance Matrix Estimation Procedures , 1995 .

[25]  Adam Groce,et al.  Differentially Private Nonparametric Hypothesis Testing , 2019, CCS.

[26]  A. C. Davison,et al.  Statistical models: Name Index , 2003 .

[27]  Abhradeep Thakurta,et al.  Statistically Valid Inferences from Privacy-Protected Data , 2023, American Political Science Review.

[28]  Daniel Sheldon,et al.  Differentially Private Bayesian Inference for Exponential Families , 2018, NeurIPS.

[29]  Daniel Sheldon,et al.  Differentially Private Bayesian Linear Regression , 2019, NeurIPS.

[30]  B. Efron.  Nonparametric standard errors and confidence intervals , 1981 .

[31]  B. Efron.  Nonparametric estimates of standard error: The jackknife, the bootstrap and other methods , 1981 .

[32]  James R. Foulds,et al.  On the Theory and Practice of Privacy-Preserving Bayesian Data Analysis , 2016, UAI.

[33]  Andrew Bray,et al.  Differentially Private Confidence Intervals , 2020, ArXiv.

[34]  Adam D. Smith,et al.  Privacy-preserving statistical estimation with optimal convergence rates , 2011, STOC '11.

[35]  C. Geyer.  Supplementary Material for "Asymptotics of Maximum Likelihood without the LLN or CLT or Sample Size Going to Infinity" , 2005, 1206.4762.

[36]  Robert Tibshirani,et al.  Bootstrap Methods for Standard Errors, Confidence Intervals, and Other Measures of Statistical Accuracy , 1986 .

[37]  Vishesh Karwa,et al.  Finite Sample Differentially Private Confidence Intervals , 2017, ITCS.

[38]  Aleksandra B. Slavkovic,et al.  Differential Privacy for Clinical Trial Data: Preliminary Evaluations , 2009, 2009 IEEE International Conference on Data Mining Workshops.