Data Privacy and Trustworthy Machine Learning

TL;DR

This paper discusses the challenges and tradeoffs involved in maintaining data privacy while ensuring trustworthy machine learning, focusing on fairness, robustness, and explainability.

Contribution

It provides an analysis of privacy risks in machine learning and explores the balance between data privacy and other trustworthiness goals.

Findings

Identifies key privacy risks in machine learning models.

Highlights the tradeoffs between privacy and model fairness.

Emphasizes the importance of balancing privacy with robustness and explainability.

Abstract

The privacy risks of machine learning models is a major concern when training them on sensitive and personal data. We discuss the tradeoffs between data privacy and the remaining goals of trustworthy machine learning (notably, fairness, robustness, and explainability).