A Novel Review of Stability Techniques for Improved Privacy-Preserving Machine Learning

TL;DR

This paper reviews techniques to improve stability in privacy-preserving machine learning, aiming to reduce noise addition and maintain model performance while safeguarding data privacy.

Contribution

It provides a comprehensive review of stability techniques that can lessen the impact of privacy noise in machine learning models.

Findings

Stability techniques can reduce the amount of noise needed for privacy.

Enhanced stability leads to better model performance under privacy constraints.

The review identifies promising methods for future research in privacy-preserving ML.

Abstract

Machine learning models have recently enjoyed a significant increase in size and popularity. However, this growth has created concerns about dataset privacy. To counteract data leakage, various privacy frameworks guarantee that the output of machine learning models does not compromise their training data. However, this privatization comes at a cost by adding random noise to the training process, which reduces model performance. By making models more resistant to small changes in input and thus more stable, the necessary amount of noise can be decreased while still protecting privacy. This paper investigates various techniques to enhance stability, thereby minimizing the negative effects of privatization in machine learning.