CS 410 Top: Introduction to Privacy-aware Computing

Overview

The rise of Big Data and Machine Learning/Artificial Intelligence has made it critical for cybersecurity to consider not only the security challenges, such as unauthorized access, but also the privacy concerns, such as unwanted inferences. This course examines the fundamentals of data privacy and looks at the challenges and opportunities of incorporating privacy into computing.

In this course, we will explore the following topics:

1. Motivations of privacy-aware computing
2. User centered approaches: privacy Policies and access control
3. Anonymization techniques: k-anonymity, l-diversity etc
4. Randomization techniques: differential privacy, geo-Indistinguishability
5. Legal aspects of privacy: privacy regulations, compliance checks
6. (Time permitting) Privacy-preserving ML: privacy attacks on ML models, federated learning

Learning Objectives

Learning Objectives:

• Describe various kinds of challenges to protecting privacy of individuals data
• Learn about different approaches to privacy-aware computing, encompassing the what, how, and when of using these techniques.
• Reason about trade-offs between privacy and other goals (e.g., utility, usability) of computing
• Learn about privacy regulations and how to check for compliance
• Examine the complexities of data privacy issues within contexts like Location-based Applications and Machine Learning

Textbook:

This course does not require a formal textbook. Instead, the course readings will be derived from online articles, seminal research papers, and other relevant sources. The syllabus and slides will offer both required and supplementary reading resources for this class.

Grade:

There are no exams in this course. The final grade will be determined based on assignments, paper presentations, final projects, and class participation.