What is Differential Privacy?

Abstract – The problem of statistical disclosure control – revealing accurate statistics about a population while preserving the privacy of individuals – has a venerable history. An extensive literature spans multiple disciplines: statistics, theoretical computer science, security, and databases. Yet privacy breaches abound, both on paper and in practice.

This talk describes a large body of work revisiting the problem from the perspective of modern cryptography. We define differential privacy, the first mathematically rigorous and comprehensive notion of privacy tailored to private data analysis. We then present general techniques for achieving differential privacy while simultaneously preserving utility of the data, together with impossibility results that guided its development.

Finally, we describe two of the exciting new directions this work has recently taken.

Biography – Cynthia Dwork is a computer scientist at Microsoft Research who works on distributed computing, cryptography and e-mail spam prevention. She received the Dijkstra Prize in 2007 for her work on consensus problems. Dr. Dwork was elected as a Fellow of the American Academy of Arts and Sciences (AAAS) in 2008, and as a member of the National Academy of Engineering in 2008. Dr. Dwork received her Ph.D. from Cornell University in 1983.