Secure Quantized Training for Deep Learning

07/01/2021

∙

by   Marcel Keller, et al.

∙

0

∙

share

We have implemented training of neural networks in secure multi-party computation (MPC) using quantization commonly used in the said setting. To the best of our knowledge, we are the first to present an MNIST classifier purely trained in MPC that comes within 0.2 percent of the accuracy of the same convolutional neural network trained via plaintext computation. More concretely, we have trained a network with two convolution and two dense layers to 99.2 (under one hour for 99