Faster sMPC implementation

Open mhluongo opened this issue 8 years ago • 1 comments

We'll likely use https://eprint.iacr.org/2017/189.pdf, but MASCOT is also on the table

Sep 08 '17 05:09 mhluongo

After discussing in-depth with our advisors, it's difficult to imagine not implementing at least two protocols- one for arithmetic and one for boolean circuits- or more based on specialized functions. For example, we could use a modified version of MASCOT by default, and provide Xiao's work as an option. Or MASCOT for RSA, Xiao's work for AES, and another general purpose protocol.

MASCOT has a publicly available implementation, which certainly helps.

Sep 13 '17 15:09 mhluongo