English

Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks

Machine Learning 2021-12-01 v5 Artificial Intelligence Computer Vision and Pattern Recognition Neural and Evolutionary Computing Machine Learning

Abstract

In recent years, graph neural networks (GNNs) have emerged as a powerful neural architecture to learn vector representations of nodes and graphs in a supervised, end-to-end fashion. Up to now, GNNs have only been evaluated empirically -- showing promising results. The following work investigates GNNs from a theoretical point of view and relates them to the 11-dimensional Weisfeiler-Leman graph isomorphism heuristic (11-WL). We show that GNNs have the same expressiveness as the 11-WL in terms of distinguishing non-isomorphic (sub-)graphs. Hence, both algorithms also have the same shortcomings. Based on this, we propose a generalization of GNNs, so-called kk-dimensional GNNs (kk-GNNs), which can take higher-order graph structures at multiple scales into account. These higher-order structures play an essential role in the characterization of social networks and molecule graphs. Our experimental evaluation confirms our theoretical findings as well as confirms that higher-order information is useful in the task of graph classification and regression.

Keywords

Cite

@article{arxiv.1810.02244,
  title  = {Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks},
  author = {Christopher Morris and Martin Ritzert and Matthias Fey and William L. Hamilton and Jan Eric Lenssen and Gaurav Rattan and Martin Grohe},
  journal= {arXiv preprint arXiv:1810.02244},
  year   = {2021}
}

Comments

Extended version with proofs, accepted at AAAI 2019, added units of measurement of QM9 dataset into appendix, removed results from Wu et al., 2018 due to different units