Privacy in Machine Learning

The quantification of privacy risks associated with algorithms is a core issue in data privacy, which holds immense significance for privacy experts, practitioners, and regulators. I will introduce a systematic approach to assessing the privacy risks of machine learning algorithms. I will highlight our efforts towards establishing standardized privacy auditing procedures and privacy meter tools to ensure compliance with privacy regulations. I will also explore the interconnections between this methodology and the concept of differential privacy. Moreover, I will talk about the complex task of protecting privacy in a decentralized learning setting and offer insights on preserving privacy while maintaining data utility.