English

DiffWire: Inductive Graph Rewiring via the Lov\'asz Bound

Machine Learning 2023-03-07 v3 Artificial Intelligence

Abstract

Graph Neural Networks (GNNs) have been shown to achieve competitive results to tackle graph-related tasks, such as node and graph classification, link prediction and node and graph clustering in a variety of domains. Most GNNs use a message passing framework and hence are called MPNNs. Despite their promising results, MPNNs have been reported to suffer from over-smoothing, over-squashing and under-reaching. Graph rewiring and graph pooling have been proposed in the literature as solutions to address these limitations. However, most state-of-the-art graph rewiring methods fail to preserve the global topology of the graph, are neither differentiable nor inductive, and require the tuning of hyper-parameters. In this paper, we propose DiffWire, a novel framework for graph rewiring in MPNNs that is principled, fully differentiable and parameter-free by leveraging the Lov\'asz bound. The proposed approach provides a unified theory for graph rewiring by proposing two new, complementary layers in MPNNs: CT-Layer, a layer that learns the commute times and uses them as a relevance function for edge re-weighting; and GAP-Layer, a layer to optimize the spectral gap, depending on the nature of the network and the task at hand. We empirically validate the value of each of these layers separately with benchmark datasets for graph classification. We also perform preliminary studies on the use of CT-Layer for homophilic and heterophilic node classification tasks. DiffWire brings together the learnability of commute times to related definitions of curvature, opening the door to creating more expressive MPNNs.

Keywords

Cite

@article{arxiv.2206.07369,
  title  = {DiffWire: Inductive Graph Rewiring via the Lov\'asz Bound},
  author = {Adrian Arnaiz-Rodriguez and Ahmed Begga and Francisco Escolano and Nuria Oliver},
  journal= {arXiv preprint arXiv:2206.07369},
  year   = {2023}
}

Comments

27 pages, 24 figures and 6 tables. Accepted at Learning on Graphs Conference 2022. A. Arnaiz-Rodriguez et al., DiffWire: Inductive Graph Rewiring via the Lov\'asz Bound. Proceedings of the First Learning on Graphs Conference (LoG 2022), PMLR 198, Virtual Event, December, 2022

R2 v1 2026-06-24T11:51:59.205Z