I'm looking for people who: - Have advanced knowledge of cryptography, preferably including homomorphic encryption of convolutional neural nets - Possess sincere accountable commitments to post singularity problems of governance, epistemology, and ethics - Want to turn superior people into dragons DM me if that's you
Are we really there in terms of compute that homomorphic encryption is viable for CNNs?
Once in a blue moon I’ll run into a cast here where I have no idea what most of the words mean, but would like to! 😂 Well I understand dragons!
Are you looking to hire or collab? If latter, @yassinelanda and I are working on the 1st above. Happy to DM with more details if you follow back.
Does it have to be a CNN? How about using something like weighted NN that acts as a lookup table to increase ease the load of computation for the HE part?
Also this seems interesting: The paper proposes automated adaption of convolutional neural networks for evaluation over fully homomorphic encryption (AutoFHE) https://openreview.net/forum?id=Hq16Jk2bVlp