arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.11199cs.LGcs.AI
收藏

图酉消息传递

Graph Unitary Message Passing

Department of Electrical Engineering, Tsinghua University(清华大学电子工程系);Beijing National Research Center for Information Science and Technology(北京信息科学与技术国家研究中心);State Key laboratory of Space Network and Communications(空间网络与通信国家重点实验室)

Haiquan Qiu, Quanming Yao

AI总结:

针对图神经网络中不稳定问题,提出图酉消息传递框架,通过在变换图上用酉传播算子避免指数衰减,结合图变换和酉投影程序,经理论分析表明其能保持深度稳定,实验取得较好性能。

摘要:

酉性是稳定深度神经网络的有用原则,但在图神经网络(GNN)中,不稳定性不仅由可学习参数引起,还由图传播算子引起。基于此区别,我们提出图酉消息传递(GUMP),一种消息传递框架,它在变换图上使用酉传播算子以避免在重复传播下由图引起的指数衰减。GUMP结合了(i)将输入图映射到允许酉邻接矩阵的欧拉图线图构造的图变换,以及(ii)基于牛顿 - 舒尔茨迭代的实用酉投影程序。理论分析表明,在标准分析假设下,酉传播使图传播项深度稳定,而普通归一化传播在其非平凡谱分量中表现出指数衰减。在合成远程任务、TUDataset基准和LRGB数据集中,GUMP优于普通消息传递,并在与强大基线的比较中实现了有竞争力或更优的性能。代码可在此https URL获取。

英文摘要:

Unitarity is a useful principle for stabilizing deep neural networks, but in graph neural networks (GNNs) instability is induced not only by learnable parameters but also by the graph propagation operator. Motivated by this distinction, we propose Graph Unitary Message Passing (GUMP), a message-passing framework that uses a unitary propagation operator on a transformed graph to avoid graph-induced exponential decay under repeated propagation. GUMP combines (i) a graph transformation that maps an input graph to an Eulerian line-graph construction admitting unitary adjacency matrices, and (ii) a practical unitary projection procedure based on Newton-Schulz iteration. Theoretical analysis clarifies that, under standard analysis assumptions, unitary propagation keeps the graph-propagation term depth-stable, while vanilla normalized propagation exhibits exponential decay in its non-trivial spectral components. Across synthetic long-range tasks, TUDataset benchmarks, and LRGB datasets, GUMP improves over vanilla message passing and achieves competitive or superior performance against strong baselines. Code is available at https://github.com/ucker/gump_code.

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。

↑