Privacy-Preserving Deep Learning With Homomorphic Encryption: An Introduction

Privacy-preserving deep learning with homomorphic encryption (HE) is a novel and promising research area aimed at designing solutions that operate while guaranteeing the privacy of user data. Designing privacy-preserving requires one to completely rethink redesign models algorithms match severe technological algorithmic constraints HE. This paper provides an introduction this complex as well methodology for convolutional neural networks (CNNs). was applied design version well-known LeNet-1 CNN, which successfully operated on two benchmark datasets image classification. Furthermore, details comments challenges software resources available