Finite Sample Differentially Private Confidence Intervals

We study the problem of estimating finite sample confidence intervals of the mean of a normal population under the constraint of differential privacy. We consider both the known and unknown variance cases and construct differentially private algorithms to estimate confidence intervals. Crucially, our algorithms guarantee a finite sample coverage, as opposed to an asymptotic coverage. Unlike most previous differentially private algorithms, we do not require the domain of the samples to be bounded. We also prove lower bounds on the expected size of any differentially private confidence set showing that our the parameters are optimal up to polylogarithmic factors.

[1]  John C. Duchi,et al.  Privacy and Statistical Risk: Formalisms and Minimax Bounds , 2014, ArXiv.

[2]  Martin J. Wainwright,et al.  Local Privacy and Minimax Bounds: Sharp Rates for Probability Estimation , 2013, NIPS.

[3]  Marco Gaboardi,et al.  PSI (Ψ): a Private data Sharing Interface , 2016, ArXiv.

[4]  Cynthia Dwork,et al.  Practical privacy: the SuLQ framework , 2005, PODS.

[5]  Thomas Steinke,et al.  Between Pure and Approximate Differential Privacy , 2015, J. Priv. Confidentiality.

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

[7]  Yue Wang,et al.  Differentially Private Hypothesis Testing, Revisited , 2015, ArXiv.

[8]  Stephen E. Fienberg,et al.  Differential Privacy and the Risk-Utility Tradeoff for Multi-dimensional Contingency Tables , 2010, Privacy in Statistical Databases.

[9]  Larry A. Wasserman,et al.  Differential privacy for functions and functional data , 2012, J. Mach. Learn. Res..

[10]  Daniel Kifer,et al.  A New Class of Private Chi-Square Tests , 2016, ArXiv.

[11]  Larry A. Wasserman,et al.  Random Differential Privacy , 2011, J. Priv. Confidentiality.

[12]  Kunal Talwar,et al.  On the geometry of differential privacy , 2009, STOC '10.

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

[14]  B. Bolch,et al.  The Teacher's Corner: More on Unbiased Estimation of the Standard Deviation , 1968 .

[15]  P. Massart,et al.  Adaptive estimation of a quadratic functional by model selection , 2000 .

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

[17]  Aleksandra B. Slavkovic,et al.  Differentially Private Exponential Random Graphs , 2014, Privacy in Statistical Databases.

[18]  Cynthia Dwork,et al.  Differential privacy and robust statistics , 2009, STOC '09.

[19]  Sofya Raskhodnikova,et al.  What Can We Learn Privately? , 2008, 2008 49th Annual IEEE Symposium on Foundations of Computer Science.

[20]  Constantinos Daskalakis,et al.  Priv'IT: Private and Sample Efficient Identity Testing , 2017, ICML.

[21]  P. Bickel.  Minimax Estimation of the Mean of a Normal Distribution when the Parameter Space is Restricted , 1981 .

[22]  Kobbi Nissim,et al.  Differentially Private Release and Learning of Threshold Functions , 2015, 2015 IEEE 56th Annual Symposium on Foundations of Computer Science.

[23]  Daniel Kifer,et al.  Revisiting Differentially Private Hypothesis Tests for Categorical Data , 2015 .

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

[25]  Andrew P. Soms,et al.  An Asymptotic Expansion for the Tail Area of the t -Distribution , 1976 .

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

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

[28]  A. Dasgupta.  Asymptotic Theory of Statistics and Probability , 2008 .

[29]  Aleksandra B. Slavkovic,et al.  Differentially Private Graphical Degree Sequences and Synthetic Graphs , 2012, Privacy in Statistical Databases.

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

[31]  G. Casella,et al.  Estimating a Bounded Normal Mean , 1981 .

[32]  Vishesh Karwa,et al.  Finite Sample Differentially Private Confidence Intervals ( Extended Abstract ) ∗ , 2017 .

[33]  Amos Beimel,et al.  Bounds on the Sample Complexity for Private Learning and Private Data Release , 2010, TCC.

[34]  Somesh Jha,et al.  Privacy in Pharmacogenetics: An End-to-End Case Study of Personalized Warfarin Dosing , 2014, USENIX Security Symposium.

[35]  P. Stark,et al.  Minimax expected measure confidence sets for restricted location parameters , 2005 .

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

[37]  Larry Wasserman.  Minimaxity, Statistical Thinking and Differential Privacy , 2012, J. Priv. Confidentiality.

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

[39]  John W. Pratt,et al.  Shorter Confidence Intervals for the Mean of a Normal Distribution with Known Variance , 1963 .

[40]  Kobbi Nissim,et al.  Simultaneous Private Learning of Multiple Concepts , 2015, ITCS.

[41]  Stephen E. Fienberg,et al.  Testing Statistical Hypotheses , 2005 .

[42]  Ashwin Machanavajjhala,et al.  Principled Evaluation of Differentially Private Algorithms using DPBench , 2015, SIGMOD Conference.

[43]  Aleksandra B. Slavkovic,et al.  Private Posterior distributions from Variational approximations , 2015, ArXiv.