arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.06706cs.CGcs.LG
收藏

CoRe-GD:基于GNN的可扩展图可视化层次化框架

CoRe-GD: A Hierarchical Framework for Scalable Graph Visualization with GNNs

ETH Zurich(苏黎世联邦理工学院)

Florian Grötschla, Joël Mathys, Robert Veres, Roger Wattenhofer

AI总结:

提出基于GNN的层次化图绘制框架CoRe-GD,通过粗化层次结构与位置重连技术,在次二次运行时间下学习优化Stress,实现可扩展的最先进性能。

摘要:

图可视化,又称图绘制,旨在寻找优化特定准则的图的几何嵌入。Stress是一种广泛使用的度量指标;当每对节点被定位在其最短路径距离时,Stress达到最小。然而,由于其固有的复杂性,Stress优化带来了计算挑战,在实践中通常使用启发式方法求解。我们引入了一个基于图神经网络(GNN)的可扩展图绘制框架,具有次二次运行时间,能够学习优化Stress。受经典的Stress优化技术和力导向布局算法启发,我们为输入图创建了一个粗化层次结构。从最粗级别开始,我们迭代地细化和解粗化布局,直到为原始图生成嵌入。为了增强网络内的信息传播,我们提出了一种基于中间节点位置的新型位置重连技术。我们的实证评估表明,该框架在保持可扩展性的同时实现了最先进的性能。

英文摘要:

Graph Visualization, also known as Graph Drawing, aims to find geometric embeddings of graphs that optimize certain criteria. Stress is a widely used metric; stress is minimized when every pair of nodes is positioned at their shortest path distance. However, stress optimization presents computational challenges due to its inherent complexity and is usually solved using heuristics in practice. We introduce a scalable Graph Neural Network (GNN) based Graph Drawing framework with sub-quadratic runtime that can learn to optimize stress. Inspired by classical stress optimization techniques and force-directed layout algorithms, we create a coarsening hierarchy for the input graph. Beginning at the coarsest level, we iteratively refine and un-coarsen the layout, until we generate an embedding for the original graph. To enhance information propagation within the network, we propose a novel positional rewiring technique based on intermediate node positions. Our empirical evaluation demonstrates that the framework achieves state-of-the-art performance while remaining scalable.

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。

↑