提出一种高效可扩展的图压缩方法，保持结构特征并提升通用性。
An Efficient and Scalable Graph Condensation with Structure-Preserving
- 分离节点压缩与结构生成，避免耦合优化
- 通过热核传播和混合聚类提取关键节点特征
- 使用预训练边预测器生成可迁移结构，适配多种GNN
图压缩（GC）对于在资源受限场景中部署图神经网络（GNN）至关重要，可通过将大规模图压缩为紧凑的合成图实现。现有方法常因耦合优化导致计算效率低下，并在不同GNN架构间泛化能力差。为此，本文提出一种高效可扩展的结构保持图压缩方法（SP-ESGC），采用解耦设计，将节点压缩与图结构生成分离。首先，基于谱图理论的热核特征传播生成节点表示；其次，设计新型混合聚类策略从节点表示中提取具有判别性的类内中心点；最后，利用预训练边预测器从原始图中推断可迁移的结构模式，确保合成图生成精度。在真实世界图数据集上的大量实验表明，SP-ESGC 实现了高精度图压缩且计算效率显著提升，同时在多种GNN架构间表现出良好泛化能力。
原文摘要 · Abstract (English)
Graph condensation (GC) is pivotal for enabling Graph Neural Networks (GNNs) deployment in resource-constrained scenarios by compressing large-scale graphs into compact synthetic counterparts. Existing GC methods commonly suffer from computational inefficiency due to coupled optimization as well as encountering poor generalization across GNN architectures. To address these challenges, this study proposes an Efficient and Scalable Graph Condensation with Structure-Preserving (SP-ESGC), which possesses a decoupled design that separates node condensation from graph structure generation. Specifically, it first employs heat kernel feature propagation to generate node representation via spectral graph theory-inspired diffusion. Further, a novel hybrid clustering strategy is designed to extracts discriminative intra-class centroids from the node representation. Finally, a pre-trained edge predictor infers transferable structural patterns from the original graph, ensuring accurate synthetic graph generation. Extensive experiments on real-world graph datasets demonstrate that the proposed SP-ESGC implementes a precise GC with significantly high computational efficiency. Moreover, SP-ESGC also generalizes well across diverse GNN architectures.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。