A07 Signal processing on graphs and complexes

The goal of this project is to advance our understanding of graph neural networks and related signal processing methods for data defined on non-Euclidean domains. We will explore two selected facets within this broad area: a) how the structure of the domain on which the neural network is defined constrains its expressibility, and b) how the structure of the neural network influences its training via gradient descent and related algorithms. In both cases we concentrate on the roles of symmetries and (approximate) low rank structures within those networks, i.e., low-dimensional (sparse) substructures within these networks.

Project Leader
Postdoctoral Researchers

Publications

  • A07
    D. Patel, A. Savostianov, M. T. Schaub

    Convergence of gradient based training for linear Graph Neural Networks

    Preprint 2025

    bibtex publications.rwth-aachen.de doi.org

  • A07
    B. Epping, A. René, M. Helias, M. T. Schaub

    Graph Neural Networks Do Not Always Oversmooth

    Preprint 2024

    bibtex publications.rwth-aachen.de doi.org