Week 1
Aug 24-28
Monday
No class. Arrival and welcome period.
Wednesday
Course overview and what privacy means in practice.
No deliverable
How can we use data to build useful systems without exposing the people behind the data? This course introduces the core ideas of modern data privacy through concrete attacks, practical defenses, and hands-on labs.
The course is designed for advanced undergraduates. We will start with privacy failures that students can observe directly, then build toward differential privacy, privacy-aware machine learning, and privacy-enhancing technologies such as MPC, HE, TEE, and network privacy tools. The emphasis is on technical understanding, experimental reasoning, and clear communication rather than graduate-level novelty.
This version of the course is aimed at advanced undergraduates in computer science, data science, or related areas.
Required background:
Recommended background:
You do not need prior experience with LLM training, privacy research, or advanced cryptography.
Privacy is now part of the job in machine learning, data science, and systems work. Engineers are expected to understand not only how to build models, but also how those models leak, what protections are realistic, and where the trade-offs appear in practice. This course is intended to prepare students for that level of technical judgment.
Tentative Monday/Wednesday plan for Fall 2026. It follows the UVA academic calendar: courses begin on August 25, Fall Reading Days run October 3-6, Thanksgiving recess runs November 25-29, and courses end on December 8. Room assignments and some due dates may still change.
No class. Arrival and welcome period.
Course overview and what privacy means in practice.
ML background for privacy.
Privacy attacks: extraction and memorization.
Attack recap and extraction discussion.
Membership inference and attack evaluation.
Linkage, singling-out, and reconstruction.
Defenses before DP: anonymization and its limits.
Differential privacy: definition, adjacency, sensitivity.
Laplace, Gaussian, and report noisy max.
Composition and privacy accounting.
DP case studies in data analysis.
No class. Fall Reading Days.
Private learning: DP-SGD intuition and practice.
Local DP and federated settings.
Exponential mechanism and private selection.
Project workshop and paper discussion.
Cryptography background for privacy engineers.
MPC basics and trust models.
MPC for simple analytics and inference.
HE, TEE, and system trade-offs.
Network privacy, telemetry, and metadata.
Privacy engineering case studies.
Applied PETs: choosing the right tool.
Project workshop and poster clinic.
Guest lecture or advanced topic.
Review and synthesis.
No class. Thanksgiving recess.
Poster / demo session, group 1.
Poster / demo session, group 2.
Course wrap-up and what to do next in privacy.
No class. Finals period.