Auditing Data Privacy for Machine Learning

This course explains the distinction between confidentiality and privacy in machine learning, focusing on membership inference attacks and quantitative methods for auditing information leakage from model predictions and parameters. It also discusses privacy risks in large models, machine learning services, federated learning, and defenses such as differential privacy.

Syllabus

Intro
Main Takeaways . There is a difference between confidentiality and privacy
Privacy Regulations
Indirect Privacy Risks in Machine Learning
Machine Learning as a Service Platforms
Large Language Models
Federated Learning Algorithms
Membership Inference Attack
Al Regulations and Guidelines
Example: Language Generative Model
Examples of Vulnerable Training Data
Example: Image Classification Tasks
Auditing Data Privacy for Machine Learning