Eastern Great Lakes Theory Workshop Talk

Integrating Differential Privacy with Statistical Theory

Adam Smith, Pennsylvania State University

Sunday, October 4, 10:30am-11:30am

ABSTRACT

Differential privacy is a definition of privacy in statistical databases that that provides meaningful confidentiality guarantees in the presence of arbitrary side information. In this talk, we describe recent work on the asymptotic properties of differentially private statistical estimators. For several problems, we design methods that are asymptotically efficient, meaning that they produce answers that converge to the correct value at the same rate as the optimal nonprivate estimator (as the number of sample points increases). These results show a range of settings in which privacy is achieved at no asymptotic cost. They provide further evidence that rigorous definitions of privacy are compatible with valid statistical inference.

Slides

Speaker Bio

Adam Smith is an assistant professor in the Department of Computer Science and Engineering at Penn State. His research interests lie in cryptography, privacy and their connections to information theory, quantum computing and statistics. He received his Ph.D. from MIT in 2004 and was subsequently a visiting scholar at the Weizmann Institute of Science and UCLA.

< Schedule < EaGL-II homepage