The goal of secure function computation is for mutually distrusting parties in a network to collaborate without relying on a trusted server in computing a function of their data in such a way they end up learning no more information about each other's data than they can infer from their own data and the function value. In other words, the parties want to simulate the presence of a trusted server which could compute the function and only provide the answer to the parties. I will describe some of our recent results on showing optimality/impossibility results for communication requirements to securely compute. The main tools we introduce include a new class of information inequalities for networks and a generalization of the notions of common information of Gács & Körner (1973) and Wyner (1975). (Joint work with Deepesh Data (TIFR) and Manoj Prabhakaran (UIUC))
Vinod M. Prabhakaran received the ME degree from the IISc Bangalore in 2001 and the PhD degree from the University of California, Berkeley in 2007. He was a Postdoctoral Researcher at the Coordinated Science Laboratory, UIUC from 2008 to 2010 and at EPFL Switzerland in 2011. Since 2011, he has been a Reader at the School of Technology and Computer Science at TIFR Mumbai. His research interests are in information theory, wireless communication, cryptography, and signal processing. He has received the Tong Leong Lim Pre-Doctoral Prize and the Demetri Angelakos Memorial Achievement Award from the EECS Department, University of California, Berkeley, and the Ramanujan Fellowship from the Department of Science and Technology, Government of India.