Characterization of Privacy Loss

 

December 17,  2015
11:00am - 1:00pm
EE Conference Room (Mudd 13 floor)
Speaker: Dr. Gail Gilboa-Freedman
Computer Science, Tel Aviv University and Electrical Engineering, Columbia University

Abstract

How much privacy is lost during a process? This question is not a new question. However, an un-equivocal answer is still missing in the scientific literature. Our research aims to formalize the conceptual notion of privacy. The applicability of our research is for having a standard methodology of prioritizing procedures by their level of privacy-loss. In this talk, I'll present our preference relation model over privacy jeopardizing processes. A list of 4 natural properties (axioms) describes this relation. Our Characterization Theorem identifies that f-divergence represents any preference relation model which satisfies these axioms, and therefore can serve as a natural measure of privacy-loss. Our second Theorem follows a reverse direction of this Decision-Theory-flavored approach, in order to characterize differential privacy, which is an ad-hoc standard in computer-science literature. *Joint work with Rann Smorodinsky and Kobbi Nissim

Bio

Dr. Gail Gilboa-Freedman received the B.S. degree in Mathematics and Computer Science (cum laude) from the Technion in 2001 and the M.Sc. degree in Applied Mathematics from the Technion in 2005. She received the Ph.D. degree in Mathematics from Tel Aviv University in 2011. Over the years, she has worked for Elbit, IBM Research, and City Bank. In 2014-5 she was a Postdoctoral researcher at the Technion and she is currently a Postdoctoral Research Fellow in Tel Aviv University and Columbia University.s


500 W. 120th St., Mudd 1310, New York, NY 10027    212-854-3105               
©2014 Columbia University