avatar

Katie Keegan

Student
Emory University
kkeega3 [at] emory [dot] edu

[Research]   [CV]   [Blog]

About Me

Hello!

I’m a fifth-year PhD student (wow, time flies!) in Computational Math at Emory University, where I’m supported by a DOE Computational Science Graduate Fellowship and Emory’s Women in Natural Sciences Fellowship. I am fortunate to be advised by Lars Ruthotto.

🌱 I am on the postdoctoral job market and am actively seeking opportunities beginning in Fall 2027. I would love to chat with any folks also working on geometric or constrained generative modeling!

In fall 2026, I am visiting Ben Erichson’s group at LBNL/ICSI and working on scalable scientific diffusion models.

I enjoy thinking about incorporating domain knowledge in AI/ML for science and engineering in a scalable, efficient, and mathematically sound way. Specifically, my research interests include constrained generative modeling, computational techniques for sampling relying on geometric measure theory and PDEs, and high-dimensional data analysis by way of tensor/multilinear algebra.

Some brief summaries of my work are available in this profile article and this talk at the DOE CSGF 2026 Annual Program Review.

I have been fortunate to enjoy many collaborations at various institutions, including Michael Mahoney’s group at LBNL in summer 2026, NASA’s Frontier Development Lab for solar active region forecasting with uncertainty quantification in summer 2025, and Aydın Buluç’s group at LBNL in summer 2024.

In summer 2023, January 2024, March 2025, and March 2026, I was in Tokyo at the High Performance AI Systems Research Team at RIKEN Center for Computational Science working with Mohamed Wahib.

Thank you for kindly taking the time to visit my website!

News

🎉 July 2026: Delivered my outgoing fellow talk at my final DOE CSGF Program Review. Incredibly grateful for the program staff for such an amazing fellowship experience.

🎉 June 2026: Grateful to receive a grant for supercomputing time at NERSC for ongoing research in constrained generative modeling.

🎉 May 2026: Grateful to be featured in this profile in DOE CSGF’s magazine, DEIXIS, and especially thankful to Jacob Berkowitz for his time writing the article!

🎉 April 2026: Happy to share that my paper “Manifold-Aware Perturbations for Constrained Generative Modeling” with Lars Ruthotto was selected as a spotlight paper (top 2.2%) at ICML 2026! Also grateful to be selected as a Gold Reviewer (top 25% of conference reviewers) and to receive an ICML Student Travel Grant.

Upcoming Travel

🛩️ IMSI Workshop on Statistical Foundations of Generative Modeling Chicago, IL, October 2026

🛩️ SIAM Mathematics of Data Science 2026 Conference Salt Lake City, Utah, November 2026

Recent Travel

🛩️ ICML 2026 Seoul, South Korea, July 2026

🛩️ DOE CSGF Program Review Washington, DC, July 2026

🛩️ “Computational methods for probability distributions on manifolds” workshop Paris, France, May 2026

🛩️ SPACERAISE Summer School L’Aquila, Italy, May 2026

🛩️ Visit to RIKEN High-Performance Artificial Intelligence Systems Research Team Tokyo, Japan, March 2026

🛩️ Supercomputing 2025 St. Louis, MO, November 2025

🛩️ Machine Learning for Heliophysics Conference Madrid, Spain, September 2025

🛩️ IAIFI Summer School and Workshop Boston, MA, August 2025

Publications

Other


Powered by Jekyll and Minimal Light theme.