I am an Assistant Professor in the Department of Computer Science at the University of Virginia. I received my PhD from Purdue University and my Bachelor's degree from Fudan University.
My group studies differential privacy, private machine learning, and synthetic data, with an emphasis on methods that can survive contact with real systems, real users, and real constraints.
I care about research that connects theory to deployment: algorithms with defensible guarantees, systems that remain usable, and evaluations that hold up beyond idealized benchmarks.
Research
Differential Privacy
Mechanisms, estimation, and optimization under privacy constraints, with attention to the gap between clean theory and actual deployment conditions.
Research
Private ML and Synthetic Data
Training and evaluating useful models and synthetic datasets while preserving privacy, interpretability, and downstream utility.
Research
Systems and Practice
Turning privacy ideas into workflows, demos, and software that can be used by researchers, practitioners, and students.
Start Here
Read, build, or teach from here
- Publications for papers and bibliographic data.
- Teaching for course materials and past offerings.
- Students for current group members and alumni.
- CV for a compact academic summary.
UVA DP Lab
Current students
I am especially interested in projects that connect privacy theory to deployable systems and rigorous empirical evaluation.
Selected projects and demos
DP Advisor
A custom GPT designed to help users evaluate whether differential privacy is appropriate for their use case and avoid common pitfalls. The tool does not collect personal information.
For critical decision-making settings, please consult with a human expert. Prompt and design details are available upon request.
PrivSyn
An interactive demo that walks users through generating synthetic tabular data with differential privacy guarantees. Currently tested on the Adult dataset.
Availability may vary due to free-tier cloud hosting constraints.
UVA DP Lab
The lab maintains code, papers, demos, and shared group infrastructure across personal repositories and organization repositories. The organization page is the public entry point for shared lab resources.
More about the group: Students
Personal
My wife, Xuejun Zhao, is an Assistant Professor in Information Systems and Operations Management at UNC Charlotte.
Last updated: Mar 19, 2026