AI Lab logo
menu MENU

Faculty Candidate Seminar

New Notions of Security

Dr. Manoj Prabhakaran
SHARE:

Dr. Prabhakaran is from Princeton University
Secure multi-party computation (MPC, for short) is a powerful
cryptographic concept which lets mutually distrusting parties
collaborate without compromising their private information (beyond
what is required by the functionality for which they collaborate).
The functionality allowed in such a collaboration is so general
that MPC subsumes virtually all other cryptographic tasks. Much of
the two and a half decades of cryptographic research can be seen as
striving towards the Holy Grail of realizing secure MPC in the most
challenging scenario in which the parties carry out multiple tasks
concurrently, the entire network is adversarially controlled and
there are no universally trusted entities. In this talk we show how
to achieve this.

Prior work had failed to achieve this goal for a good reason: it
was proven to be impossible under the previously used framework for
security definitions. In this talk, we introduce a new definitional
framework which meets the same high standards of security, but
which allows us to achieve the goal described above. This answers
a central open question in cryptography, and also opens up the
possibility of designing and analyzing practical and efficient
protocols for secure MPC.

Manoj Prabhakaran is finishing his Ph.D. at Princeton University under
the supervision of Professor Amit Sahai, and is supported by an IBM
Ph.D. Fellowship. His research interests are in Theoretical Computer
Science, especially Cryptography. He graduated in 2000 with the
Institute Gold Medal and a Bachelor of Technology degree in Computer
Science and Engineering, from the Indian Institute of Technology,
Bombay.

Sponsored by

CSE Division