English

Higher-Order GNNs Meet Efficiency: Sparse Sobolev Graph Neural Networks

Machine Learning 2024-11-08 v1 Signal Processing

Abstract

Graph Neural Networks (GNNs) have shown great promise in modeling relationships between nodes in a graph, but capturing higher-order relationships remains a challenge for large-scale networks. Previous studies have primarily attempted to utilize the information from higher-order neighbors in the graph, involving the incorporation of powers of the shift operator, such as the graph Laplacian or adjacency matrix. This approach comes with a trade-off in terms of increased computational and memory demands. Relying on graph spectral theory, we make a fundamental observation: the regular and the Hadamard power of the Laplacian matrix behave similarly in the spectrum. This observation has significant implications for capturing higher-order information in GNNs for various tasks such as node classification and semi-supervised learning. Consequently, we propose a novel graph convolutional operator based on the sparse Sobolev norm of graph signals. Our approach, known as Sparse Sobolev GNN (S2-GNN), employs Hadamard products between matrices to maintain the sparsity level in graph representations. S2-GNN utilizes a cascade of filters with increasing Hadamard powers to generate a diverse set of functions. We theoretically analyze the stability of S2-GNN to show the robustness of the model against possible graph perturbations. We also conduct a comprehensive evaluation of S2-GNN across various graph mining, semi-supervised node classification, and computer vision tasks. In particular use cases, our algorithm demonstrates competitive performance compared to state-of-the-art GNNs in terms of performance and running time.

Keywords

Cite

@article{arxiv.2411.04570,
  title  = {Higher-Order GNNs Meet Efficiency: Sparse Sobolev Graph Neural Networks},
  author = {Jhony H. Giraldo and Aref Einizade and Andjela Todorovic and Jhon A. Castro-Correa and Mohsen Badiey and Thierry Bouwmans and Fragkiskos D. Malliaros},
  journal= {arXiv preprint arXiv:2411.04570},
  year   = {2024}
}
R2 v1 2026-06-28T19:51:10.477Z