6.0 Learning Outcomes

By the end of this chapter, students will be able to:

• Determine the key concept of privacy attacks in the context of machine learning systems.
• Differentiate between various types of privacy attacks: data reconstruction, membership inference, and model extraction.
• Describe real-world examples of privacy concerns, such as Google’s use of Federated Learning.
• Apply mitigation strategies of differential privacy.
• Evaluate the limitations of existing defenses of privacy-preserving mechanisms.