סמינר מחלקתי - גייל גלבוע-פרידמן

Abstract:

How much privacy is lost during a process? This question is not a new question. However, there is no unequivocal answer in the scientific literature. Our research aims to formalize the conceptual notion of privacy.

In this talk, I’ll present two results:

1. Construction of a preference relation model over privacy jeopardizing processes. We list 4 natural properties (axioms) to describe such relation. We prove a Characterization Theorem stating that any preference relation model which satisfies these axioms is represented by f-divergence. We conclude that  f-divergence can serve as a natural measure of privacy-loss.
2. Characterization of Differential Privacy, which is an ad-hoc standard in computer-science literature. We choose 5 axioms for the purpose of justifying an existing measure. Some of the axioms are not natural in our eyes. Thus, Differential Privacy remains unjustified, at least until another construction - which is based on natural axioms - is demonstrated.

Our study leads to a recommendation on measuring privacy loss by f-divergence functions, such as KL-divergence or Hellinger-distance. The applicability of our research is for having a rigorous methodology of prioritizing procedures by their level of privacy-loss.

*Joint work with Prof. Rann Smorodinsky and Prof. Kobbi Nissim

Bio :

Gail Gilbao-Freedman received her BA degree in Mathematics and Computer Science (cum laude) from the Technion in 2001 and her M.Sc. in Applied Mathematics from the Technion in 2005. She received her Ph.D. from the department of Statistics and Operations Research in the school of Mathematical Sciences in Tel Aviv University in 2011.

Over the years she has worked for Elbit Systems, IBM Research Lab, and the Innovation Center of Citi Group. In 2014-15, she was a Postdoctoral Research Fellow at the faculty of Industrial Engineering and Management in the Technion. She is currently a Postdoctoral Research Fellow at the school of Computer Science in Tel Aviv University and the department of Electrical Engineering in Columbia University.