Closed jon-chuang closed 4 years ago
Some form of zero-knowledge proofs primitives would indeed be required when considering stronger adversarial models.
We have no plan or ongoing project in that direction for Lattigo. Its ring
package could be used for that purpose, but we have not investigated that aspect yet.
Just some resources: Succinct Non-Interactive Zero Knowledge for a von Neumann Architecture
Closing: discussion about the ZK layer with use-cases as described in #30 #32 will be taken offline.
Is this to ensure that the computation has been performed correctly before decryption occurs? Since if it is not performed correctly, couldn't this leak secrets like a model's parameters?
I may be interested in implementing this if I know where to start.