Learning on Graphs

At the Learning on Graphs group, led by Christopher Morris at the Chair of Machine Learning and Reasoning at RWTH Aachen University, we are developing theoretical and practical approaches to machine learning on graphs. Machine learning on graphs is a relatively new field in machine learning that develops methods that automatically make predictions about graph properties. Our work is characterized by examining practical methods in terms of their theoretical properties.

We are happy to work with interested and dedicated students on these problems. Refer to thesis guidelines for further information.

Group Members

Meet our team.

Christopher Morris

Professor

I am interested in both the theoretical aspects of graph learning—such as expressivity, generalization, and optimization—and its practical applications in combinatorial optimization and the sciences.

Luis Müller

PhD Student

Those SF Bay Area people want to create God using Transformers. I don't mind, as long as I'm the one who created the Transformer.

Chendi Qian

PhD Student

I am interested in graph machine learning, particularly in modeling long-range interactions and applying GNNs to optimization algorithms.

Antoine Siraudin

PhD Student

God generated the universe, now it's my turn. I'm interested in graph generative models, with a particular focus on diffusion models, and their practical applications, such as combinatorial optimization.

Antonis Vasileiou

PhD Student

I am interested in theoretical machine learning, focusing on generalization properties of graph neural networks, as well as graph similarity measures and their implications.

Timo Stoll

PhD Student

TBD.

Solveig Wittig

PhD Student

I am interested in the theory of learning graphs and graphons. I come from a mathematical background. So please don't bother me with anything applied.

Erik Müller

Student Assistant

I am a student assistant.

Yusuf Sahin

Student Assistant

I am a student assistant.

Thesis Opportunities

We also offer supervision over thesis works for RWTH students.

Show Thesis Guidelines