[论文解读] Spatio-Temporal Graph Neural Networks: A Survey

Zahraa Al Sahili, Mariette Awad|arXiv (Cornell University)|Jan 25, 2023

Advanced Graph Neural Networks被引用 16

一句话总结

对时空图神经网络（ST-GNNs）的综述，涵盖在处理时变图数据中的算法、应用和未解决的挑战。

ABSTRACT

Graph Neural Networks have gained huge interest in the past few years. These powerful algorithms expanded deep learning models to non-Euclidean space and were able to achieve state of art performance in various applications including recommender systems and social networks. However, this performance is based on static graph structures assumption which limits the Graph Neural Networks performance when the data varies with time. Spatiotemporal Graph Neural Networks are extension of Graph Neural Networks that takes the time factor into account. Recently, various Spatiotemporal Graph Neural Network algorithms were proposed and achieved superior performance compared to other deep learning algorithms in several time dependent applications. This survey discusses interesting topics related to Spatiotemporal Graph Neural Networks, including algorithms, applications, and open challenges.

研究动机与目标

• 通过解决静态图GNN在数据随时间演变时的局限性，激发对ST-GNN的需求。
• 调查现有的ST-GNN算法，并按架构和时序建模方法进行分类。
• 突出ST-GNN在时间相关任务中相对于其他深度学习方法的关键应用优势。
• 识别开放挑战和未来方向，以引导时空图学习的研究。

提出的方法

• 基于它们如何建模空间依赖性和时间动态，对ST-GNN算法进行审查和分类。
• 讨论典型的体系结构组件，如空间图卷积和时序建模模块。
• 总结ST-GNN研究中的代表性数据集、任务和评估实践。
• 突出ST-GNN中的开放挑战和潜在研究方向。

实验结果

研究问题

• RQ1 时空图神经网络的主要算法家族有哪些？它们如何对空间和时间建模？
• RQ2 哪些应用最能从ST-GNN获益，报道的性能提升有哪些？
• RQ3 ST-GNN研究与实践当前存在哪些开放挑战与空白？

主要发现

• ST-GNNs 将传统GNN扩展为共同建模空间结构和时间演化。
• 各种ST-GNN架构将图卷积与时序建模结合，以在时变任务上实现更优的性能。
• 综述讨论了跨领域的应用并概述了该领域的开放挑战。