Search Results for author:
Found 2 papers, 0 papers with code
A methodology for training homomorphicencryption friendly neural networks
Experiments using our approach reduced the gap between the F1 score and accuracy of the models trained with ReLU and the HE-friendly model to within a mere 0. 32-5. 3 percent degradation.
HeLayers: A Tile Tensors Framework for Large Neural Networks on Encrypted Data
We present a simple and intuitive framework that abstracts the packing decision for the user.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
