Efficient Private Machine Learning by Differentiable Random Transformations

TL;DR

This paper introduces a new framework for privacy-preserving machine learning that combines random transformations with arithmetic sharing, achieving high efficiency and low computational costs suitable for large-scale data.

Contribution

It proposes a novel privacy definition and a framework that merges random transformations with arithmetic sharing to enable efficient private machine learning.

Findings

Random transformations effectively protect privacy.

The framework achieves high efficiency and low computational cost.

Suitable for large-scale machine learning tasks.

Abstract

With the increasing demands for privacy protection, many privacy-preserving machine learning systems were proposed in recent years. However, most of them cannot be put into production due to their slow training and inference speed caused by the heavy cost of homomorphic encryption and secure multiparty computation(MPC) methods. To circumvent this, I proposed a privacy definition which is suitable for large amount of data in machine learning tasks. Based on that, I showed that random transformations like linear transformation and random permutation can well protect privacy. Merging random transformations and arithmetic sharing together, I designed a framework for private machine learning with high efficiency and low computation cost.