Code Obfuscation from Encryption?
Halls department, Hall 3
Wednesday, 28 December 2016
14:00 - 15:00
Code obfuscation can be seen as "encrypting" a "program" in a way that allows us to run it, while keeping its implementation as hidden as possible. So it is natural to ask whether code obfuscation could be based on very strong recently developed forms of encryption such as: witness encryption, predicate encryption, or functional encryption. In this talk, I will describe what these primitives are, and then I will describe a set of positive and negative results that give almost a full picture for the question above.
Mohammad Mahmoudi's research focuses on the foundations of Cryptography and its interplay with Computational Complexity and Adversarial Learning. He obtained his BS degree from the Computer Engineering Department of Sharif University in 2004 and got his Ph.D. from Princeton University in 2010. He then spent a few years at Cornell University as a postdoctoral associate before joining the University of Virginia as an assistant professor in 2010, where he is currently an associate professor of Computer Science.