arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.21092cs.LGcs.AI
收藏

通过图拓扑探测图神经网络激活模式

Probing Graph Neural Network Activation Patterns Through Graph Topology

Department of Computer Science, University of Geneva, Geneva, Switzerland(日内瓦大学计算机科学系);Research Institute for Statistics and Information Science, Geneva, Switzerland(日内瓦统计与信息科学研究所);School of Engineering and Applied Sciences, Harvard University(哈佛大学工程与应用科学学院)

Floriano Tori, Lorenzo Bini, Marco Sorbi, Stéphane Marchand-Maillet, Vincent Ginis

AI总结:

本文通过分析图神经网络的激活模式,揭示了图拓扑与学习偏好之间的关系,发现全局注意力机制加剧了拓扑瓶颈并增加了负曲率的出现。

摘要:

图上的曲率概念为图拓扑提供了理论描述,突显了瓶颈和更密集连接的区域。图神经网络中信息传递范式的副产品,如过平滑和过挤压,已被归因于这些区域。然而,图拓扑与GNNs学习偏好之间的相互作用仍然不明确。通过Massive Activations,即图变换器中的极端边激活值,我们探测这种对应关系。在合成图和分子基准上,我们的发现表明,MAs并不优先集中在曲率极值处,尽管它们理论上与信息流相关。在Long Range Graph Benchmark上,我们识别出一种系统性的曲率转移:全局注意力机制加剧了拓扑瓶颈,大幅增加了负曲率的普遍性。我们的工作将曲率重新界定为一种诊断探针,用于理解何时以及为何图学习失败。

英文摘要:

Curvature notions on graphs provide a theoretical description of graph topology, highlighting bottlenecks and denser connected regions. Artifacts of the message passing paradigm in Graph Neural Networks, such as oversmoothing and oversquashing, have been attributed to these regions. However, it remains unclear how the topology of a graph interacts with the learned preferences of GNNs. Through Massive Activations, which correspond to extreme edge activation values in Graph Transformers, we probe this correspondence. Our findings on synthetic graphs and molecular benchmarks reveal that MAs do not preferentially concentrate on curvature extremes, despite their theoretical link to information flow. On the Long Range Graph Benchmark, we identify a systemic \textit{curvature shift}: global attention mechanisms exacerbate topological bottlenecks, drastically increasing the prevalence of negative curvature. Our work reframes curvature as a diagnostic probe for understanding when and why graph learning fails.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑