Postdoc, Technische Universität Wien
1 paper at NeurIPS 2025
A novel notion of local convergence, color convergence, that characterizes the random graphs on which MPNNs can be learned. A new random graph model that is universal wrt color convergence.
