Jonathan Ullman
Associate Professor
Research interests
- Privacy
- Machine learning and statistics
- Cryptography
- Algorithms
Education
- PhD in Computer Science, Harvard University
- BSE in Computer Science, Princeton University
Biography
Jonathan Ullman is an associate professor in the Khoury College of Computer Sciences at Northeastern University, based in Boston.
Ullman's research centers on the foundations of privacy for machine learning and statistics, namely differential privacy and its surprising interplay with topics such as statistical validity, robustness, cryptography, and fairness. His background is in theoretical computer science, but his work spans algorithms, cryptography, machine learning, statistics, and security. His area of teaching includes algorithms and privacy for machine learning, and he is a member of the Theory Group, the Cybersecurity and Privacy Institute, and the Institute for Experiential AI.
Ullman has been recognized with an NSF CAREER award and the Ruth and Joel Spira Outstanding Teacher Award.
Recent publications
-
Program Analysis for Adaptive Data Analysis
Citation: Jiawen Liu, Weihao Qu, Marco Gaboardi, Deepak Garg , Jonathan R. Ullman. (2024). Program Analysis for Adaptive Data Analysis Proc. ACM Program. Lang., 8, 914-938. https://doi.org/10.1145/3656414 -
How to Make the Gradients Small Privately: Improved Rates for Differentially Private Non-Convex Optimization
Citation: Andrew Lowy, Jonathan R. Ullman, Stephen J. Wright . (2024). How to Make the Gradients Small Privately: Improved Rates for Differentially Private Non-Convex Optimization ICML. https://openreview.net/forum?id=XoSF46Pc2e -
Chameleon: Increasing Label-Only Membership Leakage with Adaptive Poisoning
Citation: Harsh Chaudhari, Giorgio Severi, Alina Oprea, Jonathan R. Ullman. (2024). Chameleon: Increasing Label-Only Membership Leakage with Adaptive Poisoning ICLR. https://openreview.net/forum?id=4DoSULcfG6 -
The power of factorization mechanisms in local and central differential privacy
Citation: Alexander Edmonds, Aleksandar Nikolov, Jonathan R. Ullman. (2020). The power of factorization mechanisms in local and central differential privacy STOC, 425-438. https://doi.org/10.1145/3357713.3384297 -
Private Query Release Assisted by Public Data
Citation: Raef Bassily, Albert Cheu, Shay Moran, Aleksandar Nikolov, Jonathan R. Ullman, Zhiwei Steven Wu. (2020). Private Query Release Assisted by Public Data ICML, 695-703. http://proceedings.mlr.press/v119/bassily20a.html -
Differentially Private Fair Learning
Citation: Jagielski, Matthew, Kearns, Michael, Mao, Jieming, Oprea, Alina, Roth, Aaron, Sharifi, Saeed, & Ullman, Jonathan. (2019). Differentially Private Fair Learning. Proceedings of the 36 Th International Conference on Machine Learning. -
The Structure of Optimal Private Tests for Simple Hypotheses
Citation: Canonne, C.L., Kamath, G., McMillan, A., Smith, A.D., & Ullman, J. (2018). The structure of optimal private tests for simple hypotheses. ArXiv, abs/1811.11148. -
Distributed Differential Privacy via Shuffling
Citation: Cheu A., Smith A., Ullman J., Zeber D., Zhilyaev M. (2019) Distributed Differential Privacy via Shuffling. In: Ishai Y., Rijmen V. (eds) Advances in Cryptology – EUROCRYPT 2019. EUROCRYPT 2019. Lecture Notes in Computer Science, vol 11476. Springer, Cham -
Privately Learning High-Dimensional Distributions
Citation: Kamath, G., Li, J., Singhal, V., & Ullman, J. (2018). Privately Learning High-Dimensional Distributions. COLT. -
Tight lower bounds for differentially private selection
Citation: Thomas Steinke and Jonathan Ullman. Tight lower bounds for differentially private selection. In IEEE 58th Annual Symposium on Foundations of Computer Science (FOCS’17), 2017. -
Algorithmic stability for adaptive data analysis
Citation: Raef Bassily, Kobbi Nissim, Adam Smith, Thomas Steinke, Uri Stemmer, and Jonathan Ullman. Algorithmic stability for adaptive data analysis. In Symposium on Theory of Computing (STOC’16), 2016 -
Interactive fingerprinting codes and the hardness of preventing false discovery
Citation: Thomas Steinke and Jonathan Ullman. Interactive fingerprinting codes and the hardness of preventing false discovery. In Proceedings of The 28th Conference on Learning Theory (COLT’15), 2015 -
Robust traceability from trace amounts
Citation: Cynthia Dwork, Adam Smith, Thomas Steinke, Jonathan Ullman, and Salil Vadhan. Robust traceability from trace amounts. In IEEE 56th Annual Symposium on Foundations of Computer Science (FOCS’15), 2015.
Related News
Current PhD students
Previous PhD students
-
Lydia Zakynthinou
-
Vikrant Singhal
-
Albert Cheu