Sabrina Fowler

Sabrina Fowler

she/her

Mathematics PhD Candidate | Machine Learning & Data Science

Research

Research Interests

My research currently focuses on neural ideals. A neural ideal is a mathematical object which encodes neural firing patterns corresponding to a certain configuration of a stimulus space. My current research questions include:

What can be learned about a neural code by computing the multigraded Betti numbers of the corresponding polarized neural ideal?

It is possible to construct a neural ideal based on a deep learning model. Which properties of this ideal will give us insight into the model?

Applied Relevance

Neural ideals have the potential to provide a rigorous mathematical framework for understanding how neural networks encode and process information. A key goal of my research is to study the algebraic structure of these ideals to gain insight into deep learning model behavior – including questions of interpretability, robustness, and trustworthiness. These are critical concerns in high-stakes applications such as image classification, object detection, and pattern recognition in domains like geospatial analysis and national security.

In addition to my academic research, I work as a Research Intern at the Georgia Tech Research Institute, where I apply machine learning techniques to sponsored defense research problems including anomaly detection in network data and cross-modal retrieval systems. This combination of theoretical depth and applied experience informs both directions of my work.

Seminars

I am currently regularly attending the following seminars at University of Nebraska-Lincoln

Seminar Talks I will give this semester:

Conferences and Workshops

I served on the organizing committee for the Nebraska Conference for Undergraduate Women in Mathematics from 2023 to 2025.