Recent Publications
Papers
X. Zhou, R. Bassily. Differentially Private Worst-group Risk Minimization. ICML 2024.
R. Bassily, C. Cortes, M. Mohri, A. Mao. Differentially Private Domain Adaptation with Theoretical Guarantees. ICML 2024.
M. Menart, E. Ullah, R. Arora, R. Bassily, C. Guzman. Differentially Private Non-Convex Optimization under the KL Condition with Optimal Rates. The 35th International Conference on Algorithmic Learning Theory, ALT 2024.
R. Bassily, C. Guzman, M. Menart. Differentially Private Algorithms for the Stochastic Saddle Point Problem with Optimal Rates for the Strong Gap. COLT 2023.
R. Bassily and Z. Sun, User-level Private Stochastic Convex Optimization with Optimal Rates. ICML 2023.
R. Arora, R. Bassily, T. González, C. Guzman, M. Menart, E. Ullah, Faster Rates of Convergence to Stationary Points in Differentially Private Optimization. ICML 2023.
R. Bassily, M. Mohri, A. T. Suresh, Private Domain Adaptation from a Public Source. AISTATS 2023.
R. Arora, R. Bassily, C. Guzman, M. Menart, E. Ullah, Differentially Private Generalized Linear Models Revisited. NeurIPS 2022.
X. Zhou, R. Bassily, Task-level Differentially Private Meta Learning. NeurIPS 2022.
R. Bassily, M. Mohri, A. T. Suresh, Differentially Private Learning with Margin Guarantees, NeurIPS 2022.
R. Bassily, C. Guzman, M. Menart, Differentially Private Stochastic Optimization: New Results in Convex and Non-Convex Settings, NeurIPS 2021.
R. Bassily, C. Guzman, A. Nandi, Non-Euclidean Differentially Private Stochastic Convex Optimization, COLT 2021.
Y. Hao, S. Latif, H. Zhang, R. Bassily, A. Rountev. Differential privacy for coverage analysis of software traces. European Conference on Object-Oriented Programming (ECOOP). July 2021
R. Bassily, V. Feldman, C. Guzman, K. Talwar, Stability of SGD on Nonsmooth Convex Losses, NeurIPS 2020 - Selected for Spotlight presentation.
R. Bassily, S. Moran, A. Nandi, Learning from Mixtures of Private and Public Populations, NeurIPS 2020.
H. Zhang, Y. Hao, S. Latif, R. Bassily, A. Rountev. Differentially-Private Software Frequency Profiling under Linear Constraints, ACM SIGPLAN Conference on Systems, Programming, Languages, and Applications: Software for Humanity (SPLASH/OOPSLA), November 2020.
S. Latif, Y. Hao, H. Zhang, R. Bassily, A. Rountev. Introducing Differential Privacy Mechanisms for Mobile App Analytics of Dynamic Content, IEEE International Conference on Software Maintenance and Evolution, September 2020.
R. Bassily, A. Cheu, S. Moran, A. Nikolov, J. Ullman, Z. S. Wu, Private Query Release Assisted by Public Data, ICML 2020.
A. Nandi, R. Bassily, Privately Answering Classification Queries in the Agnostic PAC Model, ALT 2020.
H. Zhang, S. Latif, R. Bassily, A. Rountev, Differentially-Private Control-Flow Node Coverage for Software Usage Analysis. USENIX 2020.
H. Zhang, Y. Hao, S. Latif, R. Bassily, A. Rountev. A Study of Event Frequency Profiling with Differential Privacy. The 29th International Conference on Compiler Construction (CC20), San Diego, Feb. 2020.
R. Bassily, V. Feldman, K. Talwar, A. Thakurta, Private Stochastic Convex Optimization with Optimal Rates, NeurIPS 2019 - Selected for Spotlight presentation < 2.9% of total submissions.
N. Alon, R. Bassily, S. Moran, Limits of Private Learning with Access to Public Data, NeurIPS 2019.
R. Bassily, Linear Queries Estimation with Local Differential Privacy, AISTATS 2019.
R. Bassily, O. Thakkar, A. Thakurta. Model-Agnostic Private Learning, NEURIPS 2018 - Selected for oral presentation < 0.62 % of total submissions, < 3% of accepted papers).
R. Bassily, M. Belkin, S. Ma, On exponential convergence of SGD in non-convex over-parametrized learning, arXiv 1811.02564, Nov. 2018.
H. Zhang, S. Latif, R. Bassily, A. Rountev. Differentially-Private Software Analytics for Mobile Apps: Opportunities and Challenges. International Workshop on Software Analytics (SWAN’18), November 2018.
S. Ma, R. Bassily, M. Belkin. The Power of Interpolation: the Effectiveness of SGD in Modern Over-parametrized Learning, ICML 2018.
R. Bassily, S. Moran, I. Nachum, J. Shafer, and A. Yehudayoff, Learners that Leak Little Information, Algorithmic Learning Theory (ALT 2018), Lanzarote, Spain.
R. Bassily, K. Nissim, U. Stemmer, and A. Thakurta, Practical Locally Private Heavy Hitters. NIPS 2017, Long Beach, CA, Dec 2017.
R. Bassily, K. Nissim, A. Smith, T. Steinke, U. Stemmer, and J. Ullman, Algorithmic Stability for Adaptive Data Analysis. Proceedings of the 48th Annual ACM SIGACT Symposium on Theory of Computing (STOC 2016), Cambridge, MA, June 2016.
R. Bassily, Y. Freund, Typical Stability, arXiv 1604.03336, Sep. 2016.
R. Bassily and A. Smith, Local, Private, Efficient Protocols for Succinct Histograms. ACM Symposium on Theory of Computing (STOC 2015), Portland, OR, June 2015.
R. Bassily, A. Thakurta, and A. Smith, Differentially Private Empirical Risk Minimization: Efficient Algorithms and Tight Error Bounds, IEEE Symposium on Foundations of Computer Science (FOCS 2014), Philadelphia, PA, Oct. 2014.
R. Bassily, A. Groce, J. Katz, A. Smith, Coupled-Worlds Privacy: Exploiting Adversarial Uncertainty in Statistical Data Privacy, FOCS 2013, Berkeley, CA, Oct. 2013.
R. Bassily and A. Smith, Causal Erasure Channels, ACM-SIAM Symposium on Discrete Algorithms (SODA 2014), Portland, OR, Jan. 2014.
R. Bassily and S. Ulukus, Decode-and-Forward Based Strategies for Secrecy in Multiple-Relay Networks, IEEE Wireless Communications and Networking Conference, Paris, France, April 2012.
R. Bassily and S. Ulukus, Deaf Cooperation for Secrecy with a Multi-Antenna Helper, 46th Annual Conference on Information Sciences and Systems (CISS), Princeton, NJ, March 2012
R. Bassily and S. Ulukus, Deaf Cooperation for Secrecy in Multipe-Relay Networks, IEEE Globecom, Houston, TX, December 2011.
R. Bassily and S. Ulukus, Ergodic Secret Alignment for the Fading Multiple Access Wiretap Channel, IEEE International Conference on Communications, Cape Town, South Africa, May 2010.
R. Bassily and S. Ulukus, A New Achievable Ergodic Secrecy Rate Region for the Fading Multiple Access Wiretap Channel, 47th Annual Allerton Conference on Communications, Control and Computing, Monticello, IL, September 2009.
Pre-prints
R. Bassily, C. Guzman, A. Nandi, Non-Euclidean Differentially Private Stochastic Convex Optimization: Optimal Rates in Linear Time, 2022.
Journal Articles
R. Bassily, K. Nissim, A. Smith, T. Steinke, U. Stemmer, and J. Ullman, Algorithmic Stability for Adaptive Data Analysis. Invited to Special issue of SIAM Journal on Computing (SICOMP). 2022.
R. Bassily, K. Nissim, U. Stemmer, and A. Thakurta, Practical Locally Private Heavy Hitters. Journal of Machine Learning Research (JMLR), 2020.
R. Bassily, E. Ekrem, X. He, E. Tekin, J. Xie, M. Bloch, S. Ulukus, A. Yener, Cooperative Security at the Physical Layer: A Summary of Recent Advances, IEEE Signal Processing Magazine, special issue on Signal Processing for Cyber-security and Privacy, 30(5):16-28, September 2013.
R. Bassily and S. Ulukus, Deaf Cooperation and Relay Selection Strategies for Secure Communication in Multiple Relay Networks, IEEE Transactions on Signal Processing, 61(6):1544-1554, 2013.
R. Bassily and S. Ulukus, Deaf Cooperation for Secrecy with Multiple Antennas at the Helper, IEEE Transactions on Information Forensics and Security, 7(6): 1855-1864, August 2012.
R. Bassily and S. Ulukus, Secure Communication in Multiple Relay Networks Through Decode-and-Forward Strategies, Journal of Communications and Networks, special issue on Physical Layer Security, 14(4):352-363, August 2012.
R. Bassily and S. Ulukus, Ergodic Secret Alignment, IEEE Transactions on Information Theory, 58(3):1594-1611, March 2012.
Patents
K. Nissim, U. Stemmer, R. Bassily, A. Guha Thakurta, Locally Private Determination of Heavy Hitters, U.S. Patent 62/509,630.