A Survey on Deep Learning Techniques for Privacy-Preserving

Abstract

There are challenges and issues when machine learning algorithm needs to access highly sensitive data for the training process. In order to address these issues, several privacy-preserving deep learning techniques, including Secure Multi-Party Computation and Homomorphic Encryption in Neural Network have been developed. There are also several methods to modify the Neural Network, so that it can be used in privacy-preserving environment. However, there is trade-off between privacy and performance among various techniques. In this paper, we discuss state-of-the-art of Privacy-Preserving Deep Learning, evaluate all methods, compare pros and cons of each approach, and address challenges and issues in the field of privacy-preserving by deep learning.