Geometric Machine Learning

We study geometric structure in data and models and how to leverage such information

for the design of efficient machine learning algorithms with provable guarantees.

Our Research

Characterizing data geometry and learning data representations in suitable non-Euclidean spaces.

Leveraging geometric structure for efficient learning on graphs. Design of expressive and stable graph machine learning architectures.

Designing geometric architectures. Understanding learning-theoretic trade-offs in geometric settings.

Exploiting geometric structure in (non)convex (constrained) optimization.

Geometric methods for resource-efficient and interpretable scientific machine learning.

Leveraging data and model geometry for more resource-efficient and transparent foundation models.