Availability of Matrix Operations/Algorithms for Benchmarking Precision and CPU Performance for CKKS Bootstrapping

FHE Use Cases
1

Hello,
I’m looking for a way to evaluate various matrix operations using the CKKS scheme with the use of bootstrapping. I’m interested to know if the library has such functionality since I’ve already come across a paper that explores this direction titled: Cheap and Fast Iterative Matrix Inverse in Encrypted Domain, unfortunaley i have no acces to this paper. It’s not crucial that they are super efficient; I just want to evaluate them in terms of time and precision for data assessment in my master’s thesis. If anyone has any tips for me, please let me know.

Thank you very much.

2

I don’t think matrix operations have been released. See Is there a built in method to perform vector - matrix multiplication? - #2 by Caesar

1 Like
3

The functions that are exposed in OpenFHE, such as EvalSumRows and EvalSumCols, can be used to build matrix capabilities, but there are no matrix-level capabilities exposed in OpenFHE itself. There is a separate effort FHERMA that hosts FHE challenges for higher-level operations. Please see GitHub - fairmath/polycircuit: Polycircuit - an FHE components library, build via FHERMA competitions to view the solutions for prior challenges. More matrix capabilities will be added there later.