5 ms·
Theoretically, FHE can perform any computation including training neural networks. Practically, it's still not viable because of the overhead.
by ajconway 7y ago
Theoretically, FHE can perform any computation including training neural networks. Practically, it's still not viable because of the overhead.
- gumby 7y agoI don’t understand how that could work (unless you decrypted the data). The point of this encryption is that you can’t see what’s in the underlying data so two identical values will be encrypted to different values, no?
- shoo 7y agoRegard the process of fitting a statistical model as a function mapping input training data to output model parameters. If the homomorphic encryption scheme allows for arbitrary functions to be computed, it could work for particular functions of statistical model fitting (perhaps appallingly slowly). The output model will be encoded as cyphertext. It would be possible to evaluate the fitted cyphertext model on new features to make predictions, also output as cyphertext. > the library can evaluate a net-list of binary gates homomorphically at a rate of about 76 gates per second per core, without decrypting its input. It suffices to provide the sequence of gates, as well as ciphertexts of the input bits. And the library computes ciphertexts of the output bits.