Research › Data Privacy

Data Privacy and Differential Privacy

How can we learn from sensitive data without exposing the people it describes? We work on the theory and practice of differential privacy: new mechanisms for analytics and machine learning, programming languages that automatically verify that an algorithm is private, and tools that let non-experts deploy privacy correctly. Our work informs national standards through collaboration with NIST, and our usability studies ask whether real data practitioners can actually use the tools the research community has built.

Faculty

Joseph Near

Associate Professor, Department of Computer Science

Yuanyuan Feng

Assistant Professor, Department of Computer Science

Selected publications

PoPETs 2026

"We Need a Standard": Toward an Expert-Informed Privacy Label for Differential Privacy

Onyinye Dibia, Mengyi Lu, Prianka Bhattacharjee, Joseph Near, Yuanyuan Feng

Proceedings on Privacy Enhancing Technologies (PoPETs), 2026(1)
JAMIA 2026

Aim High, Stay Private: Differentially Private Synthetic Data Enables Public Release of Behavioral Health Information with High Utility

M. Ghasemizade, J. Lovato, C. Danforth, P. Dodds, L. Bloomfield, M. Price, LEMURS Team, Joseph Near

Journal of the American Medical Informatics Association (JAMIA)
PoPETs 2025

SoK: Usability Studies in Differential Privacy

Onyinye Dibia, Prianka Bhattacharjee, Brad Stenger, Steven Baldasty, Mako Bates, Ivoline Ngong, Yuanyuan Feng, Joseph Near

Proceedings on Privacy Enhancing Technologies (PoPETs), 2025(4):881–895
NAACL 2025

Differentially Private Learning Needs Better Model Initialization and Self-Distillation

Ivoline Ngong, Joseph Near, Niloofar Mireshghallah

Annual Conference of the North American Chapter of the ACL (NAACL), pp. 9009–9027
NIST 2025

Guidelines for Evaluating Differential Privacy Guarantees (NIST SP 800-226)

Joseph Near, David Darais, Naomi Lefkovitz, Gary Howarth

National Institute of Standards and Technology, Special Publication 800-226
SOUPS 2024

Evaluating the Usability of Differential Privacy Tools with Data Practitioners

Ivoline Ngong, Brad Stenger, Joseph Near, Yuanyuan Feng

USENIX Symposium on Usable Privacy and Security (SOUPS)
TOPLAS 2023

Contextual Linear Types for Differential Privacy

M. Toro, David Darais, Chike Abuah, Joseph Near, D. Arquez, F. Olmedo, É. Tanter

ACM Transactions on Programming Languages and Systems (TOPLAS)
USENIX Security 2022

Efficient Differentially Private Secure Aggregation for Federated Learning via Hardness of Learning with Errors

Timothy Stevens, Christian Skalka, Christelle Vincent, John Ring, Samuel Clark, Joseph Near

31st USENIX Security Symposium, pp. 1379–1395
USENIX Security 2022

PrivGuard: Privacy Regulation Compliance Made Easier

Lun Wang, U. Khan, Joseph Near, Q. Pang, J. Subramanian, Neel Somani, P. Gao, A. Low, Dawn Song

31st USENIX Security Symposium
OOPSLA 2019

Duet: An Expressive Higher-order Language and Linear Type System for Statically Enforcing Differential Privacy

Joseph Near, David Darais, Chike Abuah, Tim Stevens, Pranav Gaddamadugu, Lun Wang, Neel Somani, Mu Zhang, Nikhil Sharma, Alex Shan, Dawn Song

ACM SIGPLAN Conference on Object-Oriented Programming, Systems, Languages, and Applications (OOPSLA)

🏅 ACM SIGPLAN Distinguished Paper Award

All publications in this area →

Related courses

Course	Description	Instructor
CS 3110 Data Privacy Data Privacy	How to learn from sensitive data while protecting the individuals it describes. Attacks on anonymized data, k-anonymity and its limits, and the theory and practice of differential privacy, with programming assignments on real data. Prerequisites: CS 2240, CS 2250	Joseph Near
CS 5110 Advanced Data Privacy Data Privacy	Graduate treatment of data privacy: differential privacy foundations, mechanisms for analytics and machine learning, and the systems that enforce them. Taught alongside CS 3110 with additional graduate-level work.	Joseph Near

All security and privacy courses →

Funded projects

NSF

CAREER: Distributed Differential Privacy via Secure Multiparty Computation

CAREER (Secure and Trustworthy Cyberspace) · CNS-2238442 · 2023–2028 · $548,929 · UVM PI: Joseph Near
NSF

A Novel Mechanism for Effective and Accountable Communication of Differential Privacy through Multi-Stakeholder Research

SaTC: CORE: Small · CNS-2336550 · 2024–2027 · $599,994 · UVM PIs: Yuanyuan Feng, Joseph Near
IARPA

VOXLET: Vocal Obfuscation eXpert Language Encoding Technology

Anonymous Real-Time Speech (ARTS) · UVM PI: Joseph Near · Real-time speech anonymization with differential privacy guarantees; with Galois, GE Aerospace, and North Point Defense.
DARPA

Helio: Program Synthesis for Efficient, Privacy-Preserving Distributed Computation

Brandeis · 2015–2020 · UVM PI: Joseph Near
DARPA

CLAMPED: Collaborative Learning Architecture with Mathematical Privacy over Embedded Data

CSL · 2020–2022 · UVM PI: Joseph Near

Work with us

Interested in data privacy? See how to join the Center as an MS, PhD, or undergraduate researcher.