Abstract
f-divergences form a class of measures of distance between probability distributions; they are widely used in areas such as information theory and signal processing. In this paper, we unveil a new connection between f-divergences and differential privacy, a confidentiality policy that provides strong privacy guarantees for private data-mining; specifically, we observe that the notion of α-distance used to characterize approximate differential privacy is an instance of the family of f-divergences. Building on this observation, we generalize to arbitrary f-divergences the sequential composition theorem of differential privacy. Then, we propose a relational program logic to prove upper bounds for the f-divergence between two probabilistic programs. Our results allow us to revisit the foundations of differential privacy under a new light, and to pave the way for applications that use different instances of f-divergences.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Abate, A.: Approximation metrics based on probabilistic bisimulations for general state-space markov processes: a survey. Electronic Notes in Theoretical Computer Sciences (2012) (in print)
Ali, S.M., Silvey, S.D.: A general class of coefficients of divergence of one distribution from another. Journal of the Royal Statistical Society. Series B (Methodological) 28(1), 131–142 (1966)
Alvim, M.S., Andrés, M.E., Chatzikokolakis, K., Palamidessi, C.: On the relation between differential privacy and Quantitative Information Flow. In: Aceto, L., Henzinger, M., Sgall, J. (eds.) ICALP 2011, Part II. LNCS, vol. 6756, pp. 60–76. Springer, Heidelberg (2011)
Barthe, G., Grégoire, B., Heraud, S., Béguelin, S.Z.: Computer-aided security proofs for the working cryptographer. In: Rogaway, P. (ed.) CRYPTO 2011. LNCS, vol. 6841, pp. 71–90. Springer, Heidelberg (2011)
Barthe, G., Grégoire, B., Zanella-Béguelin, S.: Formal certification of code-based cryptographic proofs. In: 36th ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages, POPL 2009, pp. 90–101. ACM, New York (2009)
Barthe, G., Köpf, B., Olmedo, F., Zanella-Béguelin, S.: Probabilistic relational reasoning for differential privacy. In: 39th ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages, POPL 2012, pp. 97–110. ACM, New York (2012)
Chaudhuri, S., Gulwani, S., Lublinerman, R., Navidpour, S.: Proving programs robust. In: 19th ACM SIGSOFT Symposium on the Foundations of Software Engineering and 13rd European Software Engineering Conference, ESEC/FSE 2011, pp. 102–112. ACM, New York (2011)
Cortes, C., Mohri, M., Rastogi, A.: Lp distance and equivalence of probabilistic automata. Int. J. Found. Comput. Sci. 18(4), 761–779 (2007)
Cortes, C., Mohri, M., Rastogi, A., Riley, M.: On the computation of the relative entropy of probabilistic automata. Int. J. Found. Comput. Sci. 19(1), 219–242 (2008)
Csiszár, I.: Eine informationstheoretische ungleichung und ihre anwendung auf den beweis der ergodizitat von markoffschen ketten. Publications of the Mathematical Institute of the Hungarian Academy of Science 8, 85–108 (1963)
Deng, Y., Du, W.: Logical, metric, and algorithmic characterisations of probabilistic bisimulation. Tech. Rep. CMU-CS-11-110, Carnegie Mellon University (March 2011)
Dwork, C.: Differential privacy. In: Bugliesi, M., Preneel, B., Sassone, V., Wegener, I. (eds.) ICALP 2006. LNCS, vol. 4052, pp. 1–12. Springer, Heidelberg (2006)
Ebanks, B., Sahoo, P., Sander, W.: Characterizations of Information Measures. World Scientific (1998)
Gaboardi, M., Haeberlen, A., Hsu, J., Narayan, A., Pierce, B.C.: Linear dependent types for differential privacy. In: 40th ACM SIGPLAN–SIGACT Symposium on Principles of Programming Languages, POPL 2013, pp. 357–370. ACM, New York (2013)
Jonsson, B., Yi, W., Larsen, K.G.: Probabilistic extensions of process algebras. In: Bergstra, J., Ponse, A., Smolka, S. (eds.) Handbook of Process Algebra, pp. 685–710. Elsevier, Amsterdam (2001)
McSherry, F.: Privacy integrated queries: an extensible platform for privacy-preserving data analysis. Commun. ACM 53(9), 89–97 (2010)
McSherry, F.D.: Privacy integrated queries: an extensible platform for privacy-preserving data analysis. In: 35th SIGMOD International Conference on Management of Data, SIGMOD 2009, pp. 19–30. ACM, New York (2009)
Pardo, M., Vajda, I.: About distances of discrete distributions satisfying the data processing theorem of information theory. IEEE Transactions on Information Theory 43(4), 1288–1293 (1997)
Pierce, B.C.: Differential privacy in the programming languages community. Invited Tutorial at DIMACS Workshop on Recent Work on Differential Privacy Across Computer Science (2012)
Di Pierro, A., Hankin, C., Wiklicky, H.: Measuring the confinement of probabilistic systems. Theor. Comput. Sci. 340(1), 3–56 (2005)
Reed, J., Pierce, B.C.: Distance makes the types grow stronger: a calculus for differential privacy. In: 15th ACM SIGPLAN International Conference on Functional programming, ICFP 2010, pp. 157–168. ACM, New York (2010)
Roy, I., Setty, S.T.V., Kilzer, A., Shmatikov, V., Witchel, E.: Airavat: security and privacy for MapReduce. In: 7th USENIX Conference on Networked Systems Design and Implementation, NSDI 2010, pp. 297–312. USENIX Association, Berkeley (2010)
Segala, R., Turrini, A.: Approximated computationally bounded simulation relations for probabilistic automata. In: 20th IEEE Computer Security Foundations Symposium, CSF 2007, pp. 140–156. IEEE Computer Society (2007)
Steinberger, J.: Improved security bounds for key-alternating ciphers via hellinger distance. Cryptology ePrint Archive, Report 2012/481 (2012), http://eprint.iacr.org/
Tracol, M., Desharnais, J., Zhioua, A.: Computing distances between probabilistic automata. In: Proceedings of QAPL. EPTCS, vol. 57, pp. 148–162 (2011)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Barthe, G., Olmedo, F. (2013). Beyond Differential Privacy: Composition Theorems and Relational Logic for f-divergences between Probabilistic Programs. In: Fomin, F.V., Freivalds, R., Kwiatkowska, M., Peleg, D. (eds) Automata, Languages, and Programming. ICALP 2013. Lecture Notes in Computer Science, vol 7966. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-39212-2_8
Download citation
DOI: https://doi.org/10.1007/978-3-642-39212-2_8
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-39211-5
Online ISBN: 978-3-642-39212-2
eBook Packages: Computer ScienceComputer Science (R0)