Privacy-Preserving 3-Layer Neural Network Training

TL;DR

This paper presents a method for training 3-layer neural networks securely using homomorphic encryption, enabling privacy-preserving machine learning for regression and classification tasks.

Contribution

It introduces an approach that combines and extends existing techniques to enable privacy-preserving training of 3-layer neural networks with homomorphic encryption.

Findings

Successfully trains 3-layer neural networks with privacy guarantees

Supports both regression and classification tasks

Extends existing homomorphic encryption techniques for neural network training

Abstract

In this manuscript, we consider the problem of privacy-preserving training of neural networks in the mere homomorphic encryption setting. We combine several exsiting techniques available, extend some of them, and finally enable the training of 3-layer neural networks for both the regression and classification problems using mere homomorphic encryption technique.