Spectral Greedy Coresets for Graph Neural Networks

TL;DR

This paper introduces a spectral greedy coreset method for GNNs that efficiently selects representative ego-graphs to accelerate training on large, complex graphs, outperforming existing methods in speed and accuracy.

Contribution

The paper proposes a novel spectral greedy algorithm for graph coreset selection that effectively handles node interdependence and scales to large graphs, improving GNN training efficiency.

Findings

SGGC outperforms other coreset methods significantly.

It generalizes well across different GNN architectures.

It is much faster than graph condensation techniques.

Abstract

The ubiquity of large-scale graphs in node-classification tasks significantly hinders the real-world applications of Graph Neural Networks (GNNs). Node sampling, graph coarsening, and dataset condensation are effective strategies for enhancing data efficiency. However, owing to the interdependence of graph nodes, coreset selection, which selects subsets of the data examples, has not been successfully applied to speed up GNN training on large graphs, warranting special treatment. This paper studies graph coresets for GNNs and avoids the interdependence issue by selecting ego-graphs (i.e., neighborhood subgraphs around a node) based on their spectral embeddings. We decompose the coreset selection problem for GNNs into two phases: a coarse selection of widely spread ego graphs and a refined selection to diversify their topologies. We design a greedy algorithm that approximately optimizes both objectives. Our spectral greedy graph coreset (SGGC) scales to graphs with millions of nodes, obviates the need for model pre-training, and applies to low-homophily graphs. Extensive experiments on ten datasets demonstrate that SGGC outperforms other coreset methods by a wide margin, generalizes well across GNN architectures, and is much faster than graph condensation.