Data Privacy (Spring 2025)

Note: Under construction. This is a direct copy of the Spring 2022 offering. Everything may change dramatically. How can we learn from data while protecting individuals’ privacy? This course addresses that question, starting with privacy attacks and progressing to rigorous, state-of-the-art solutions, such as differential privacy. We will cover both theoretical foundations and practical challenges involved in real-world applications. Prerequisites for this course include a basic understanding of machine learning (e.g., knowledge of how large language models work), proficiency in coding (e.g., Python), and familiarity with reading and writing algorithmic proofs that involve probability.

Instructor: Tianhao Wang
Co-Instructor: Xiaochen Li
Location: Rice 340
Time: Monday and Wednesday 12:30pm - 1:45pm
TA: TBD
Discussion: Piazza
Office Hour
- Tianhao: TBD
Policy

Grading:

Paper presentation (20%): During the semester, each student will present two papers, each for 15 minutes. If the slides are public or there are videos available online, it is okay to reuse the slides, but make sure you make a high-quality presentation. Click here to register.
Cyber Chats (10%): Each student investigates two related topics and make recordings each of 5min.
X (X=2 or 3) assignments (20%) for theory (proofs) and practice (programming)
Quiz
Project (50%): Click here to register your project with a tentative title. Students are encouraged utilize office hours to get feedbacks on the project. Students need to submit project proposals (15%), present a poster (15%), and submit your final report (20%).
Grading Scale: We use the UVA default Grading Basis

Schedule (tentative, subject to change), we will meet in-person

Week	Dates	Monday	Wednesday
1	Jan 13 - Jan 17	Intro	Privacy Attacks
2	Jan 20 - Jan 24	Privacy Attacks	Privacy Attacks
3	Jan 27 - Jan 31	Privacy Attacks	Privacy Attacks
4	Feb 3 - Feb 7	Privacy Enhancing Technologies	Privacy Enhancing Technologies (HW 1 Due)
5	Feb 10 - Feb 14	Privacy Enhancing Technologies	Privacy Enhancing Technologies
6	Feb 17 - Feb 21	Privacy Enhancing Technologies	Privacy Enhancing Technologies
7	Feb 24 - Feb 28	Spring break	Spring break (HW 2 Due)
8	Mar 3 - Mar 7	Student-led Presentation	Student-led Presentation
9	Mar 10 - Mar 14	Student-led Presentation	Student-led Presentation
10	Mar 17 - Mar 21	Student-led Presentation	Student-led Presentation (Cyber Chat Due)
11	Mar 24 - Mar 28	Cyber Chat Discussion	Cyber Chat Discussion
12	Mar 31 - Apr 4	Student-led Presentation	Student-led Presentation (Project Proposal Due)
13	Apr 7 - Apr 11	Student-led Presentation	Student-led Presentation
14	Apr 14 - Apr 18	Student-led Presentation	Student-led Presentation
15	Apr 21 - Apr 25	Work on Poster	Work on Final Project (Final Project Poster Due)
16	Apr 28 - Apr 30	Poster Session	No Class (Final Project Report Due)

More recourses

Privacy in Statistics and Machine Learning (video) Spring 2023 by Adam Smith (BU) and Jonathan Ullman (NEU)
Algorithms for Private Data Analysis (video) Fall 2022 by Gautam Kamath (Waterloo)
Applied Privacy for Data Science Spring 2022 by James Honaker, Wanrong Zhang, and Salil Vadhan (Harvard)
Introduction to Differential Privacy: Theory, Algorithms and Applications (video) Fall 2024 by Yuxiang Wang (UCSD)
Algorithms for Private Data Analysis Fall 2020 by Aleksandar Nikolov (UofT)

Books

More courses on other flavors

Theory: The Algorithmic Foundations of Adaptive Data Analysis Fall 2017 by Aaron Roth (Penn) and Adam Smith (BU)
System: Private Systems Spring 2020 by Roxana Geambasu (Columbia)
Mechanism design: Differential Privacy in Game Theory and Mechanism Design Spring 2014 by Aaron Roth (Penn)
Fairness: Privacy and Fairness in Data Science Fall 2021 by Ashwin Machanavajjhala (Duke)
ML: Privacy Preserving Machine Learning Spring 2022 by Aurélien Bellet (Inria)