[Busec] busec next week: Manoj Prabhakaran (Wed 10am)

Sharon Goldberg goldbe at cs.bu.edu
Tue Jul 1 12:17:15 EDT 2014

I hope everyone is having a quiet and productive summer!  Manoj Prabhakaran
from UIUC will be visiting us next week, to give a seminar about a unified
theory of cryptographic agents.  Seminar will be 10am on Wed July 9, with
lunch following.

Hope to see many of you there,


 BUsec Calendar:  http://www.bu.edu/cs/busec/
 BUsec Mailing list: http://cs-mailman.bu.edu/mailman/listinfo/busec
 How to get to BU from MIT: The CT2 bus or MIT's "Boston Daytime Shuttle"

Title: Towards a Unified Theory of Cryptographic Agents
Speaker: Manoj Prabhakaran, UIUC.
July 9, 2014, 10-11:30 am
MCS137, at 111 Cummington St, Boston MA


 In recent years there has been a fantastic boom of increasingly
sophisticated "cryptographic objects" --- Identity-Based Encryption,
Fully-Homomorphic Encryption, Functional Encryption, various forms of
obfuscation, Witness Encryption, Property-Preserving Encryption, etc.
As these constructions have grown in number, variety, complexity and
inter-connectedness, the relationships among them have become
increasingly confusing.

 We provide a new framework of {\em cryptographic agents} that unifies
various cryptographic objects and security definitions, similar to how the
Universal Composition framework unifies various multi-party
computation tasks like commitment, coin-tossing and zero-knowledge
proofs. Various primitives are modeled by different "schemas"
(analogous to functionalities in the UC framework).

 We use a new {\em indistinguishability preserving} (INDPRE) definition
of security, that often side-steps the impossibility results
associated with simulation-based definitions, and yet admits a
composition theorem. Also, when appropriately restricted, our
definitions often yield existing definitions of the corresponding

 Interestingly, our framework can also be used to model abstractions
like the generic group model and the random oracle model, letting one
translate a general class of constructions in these heuristic models
to constructions based on {\em standard model assumptions}.

 We illustrate the new framework by presenting new constructions of
functional encryption (FE) schemes, with and without function-hiding,
in terms of schematic reductions to the obfuscation schema or the
bilinear generic group schema. When combined with candidate schemes for
these schemas using our composition theorem, these constructions yield
concrete FE schemes.

(Joint work with Shweta Agrawal and Shashank Agrawal)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs-mailman.bu.edu/pipermail/busec/attachments/20140701/b4e32862/attachment.html>

More information about the Busec mailing list