2018-03-16 | Siyao Guo：Non-uniform security: limits on the power of pre-computation
Cryptographic hash functions are widely used in practice for a variety of applications, including password hashing, digital signatures, bitcoin and more. The fact that hash functions are public allows attackers to perform pre-computation before attacking the system. However, the traditional security bounds for the above applications either do not apply at all, or do not give accurate security bounds when pre-computation is allowed.
In this talk, we focus on non-uniform attackers, which can obtain arbitrary (but bounded-length) pre-computed advice about the hash function before attacking the system. We introduce optimal and generic tools for analyzing the non-uniform security of hash-function-based applications. Furthermore, we give simple compilers that transform any secure scheme (in the traditional sense) into one that is secure against non-uniform attackers.
Siyao Guo is a postdoctoral fellow at the Cybersecurity and Privacy Institute, Northeastern University. Previously she was a postdoctoral fellow at the Simons Institute for the theory of computation, UC Berkeley, and the Courant Institute of Mathematical Sciences, New York University. She obtained her Ph.D. from the Chinese University of Hong Kong in 2014. Her research interests lie in cryptography, computational complexity, and pseudorandomness.