Hyemin Gu

I am a postdoctoral researcher in Markos Katsoulakis’s group at UMass Amherst, where I completed my Ph.D. in Applied Mathematics with the dissertation Robust Generative Modeling with Wasserstein Proximal Regularization. My research places machine learning models on mathematical foundations, so that their behavior can be predicted, analyzed, and controlled through the underlying theory.

Research interests. Mathematically grounded machine learning; generative modeling; stable learning algorithms; scientific applications of machine learning.

Current research directions.

• mathematically characterizing stable learning architectures for reliable training and generalization;
• developing generative modeling methods based on Wasserstein gradient flows and variational principles for \((f,\Gamma)\)-divergence; and
• developing scalable computational methods for scientific applications, including single-cell dynamics and supply-chain networks.

A recurring challenge in mathematical sciences is balancing abstraction, which enables transferability across domains, with concrete problem formulations, which often drive practical advances. My long-term goal is to let mathematical theory and real-world applications continuously inform and strengthen one another.