4 ms·
Since TFHE does the encryption at the binary level, it looks like it would need to also work on the float level to be useful for a lot of data science/machine l
by physicsyogi 6y ago
Since TFHE does the encryption at the binary level, it looks like it would need to also work on the float level to be useful for a lot of data science/machine learning use cases. I could be wrong though.
- malux85 6y agoIf you can do binary, then you can represent floating point numbers, no?
- nindanaoto 6y agoTFHE is known to be switchable between other RLWE based Homomorphic Encryption. https://eprint.iacr.org/2018/758 https://eprint.iacr.org/2018/758 Glyph applied it to accelerate deep learning, but they use integer level, not float. https://arxiv.org/abs/1911.07101 https://arxiv.org/abs/1911.07101 AFAIK, Homomorphic Encryption which supports floating point arithmetic is not known. Fixed point one is known, CKKS. https://github.com/snucrypto/HEAAN https://github.com/snucrypto/HEAAN