Hi SD59 people,
anyone up for reviewing
http://trac.sagemath.org/ticket/15915
It adds discrete Gaussian samplers over the integers, lattices + polynomial
rings (okay, those are the same as ZZ^n for our purposes) to Sage. Discrete
Gaussians are used in a lot of current crypto constructions (fully homomorphic
encryption, multilinear maps, ...) and having them readily available in Sage
would be very very nice.
At #15915 I wrote some C code for sampling over the integers + higher level
code for extending these samplers to lattices.
Writing code for distributions is annoying as changes tend to change the
output of many doctests without affecting correctness. Hence, it would be good
to have a short feedback loop for reviews.
Cheers,
Martin