Faculty Profile for Dr. Jeremiah Birrell
Dr. Jeremiah Birrell
Assistant Professor — Mathematics
MCS 457
phone: (512) 245-8018
Biography Section
Biography and Education
I obtained a PhD in Applied Mathematics from the University of Arizona and a Bachelor of Science in Physics from Brigham Young University. Before coming to Texas State University I was a postdoc in the Mathematics Department and the Computer Science Department at the University of Massachusetts Amherst.Teaching Interests
As an applied mathematician, my preferred teaching style is to intermingle a solid foundation in theory with practical algorithmic considerations, including non-trivial applications, which generally involve coding. I find that this approach amplifies students' abilities in both theoretical and practical aspects of the subject matter and makes the material more accessible and relevant to a diverse group of students with many different backgrounds and interests. My preferred courses to teach in this style are linear algebra, numerical analysis, nonlinear dynamics, and machine learning.Research Interests
I am an applied mathematician, currently working in statistical learning theory. Current and past research includes: 1) Privacy-preserving machine learning. 2) Adversarial robustness of deep learning models. 3) Design and analysis of novel algorithms for generative models. 4) Robustness bounds for quantities of interest in stochastic systems. 5) Singular limits of stochastic differential equations. 6) Numerical methods in kinetic theory.Selected Scholarly/Creative Work
- Birrell, J., Ebrahimi, R., Behnia, R., & Pacheco, J. (2024). Differentially Private Stochastic Gradient Descent with Fixed-Size Minibatches: Tighter RDP Guarantees with or without Replacement. Retrieved from https://neurips.cc/virtual/2024/poster/95041
- Birrell, J., Pantazis, Y., Dupuis, P., Rey-Bellet, L., & Katsoulakis, M. A. (2023). Function-space regularized Rényi divergences. Retrieved from https://openreview.net/forum?id=89GT-S49mGd
- Birrell, J., Katsoulakis, M. A., Rey-Bellet, L., & Zhu, W. (2022). Structure-preserving GANs. In Proceedings of Machine Learning Research (Vol. 162, pp. 1982–2020). Retrieved from https://proceedings.mlr.press/v162/birrell22a.html
- Birrell, J., Dupuis, P., Katsoulakis, M. A., Pantazis, Y., & Rey-Bellet, L. (2022). (f,Γ)-Divergences: Interpolating between f-Divergences and Integral Probability Metrics. Journal of Machine Learning Research, 23. Retrieved from https://www.jmlr.org/papers/v23/21-0100.html
- Birrell, J., Dupuis, P., Katsoulakis, M. A., Rey-Bellet, L., & Wang, J. (2021). Variational Representations and Neural Network Estimation of Rényi Divergences. SIAM Journal on Mathematics of Data Science, 3(4), 1093–1116. https://doi.org/10.1137/20m1368926
Selected Service Activities
Member
Colloquium Committee
September 23, 2024-Present
Worked at the Texas State Mathematics Department booth at the graduate student fair.
Joint Mathematics Meetings 2025
January 10, 2025-January 10, 2025
Speaker
II San Marcos Winter School
December 16, 2024-December 18, 2024