ScaleNet：有向图中的尺度不变性学习
ScaleNet: Scale Invariance Learning in Directed Graphs
AI总结：
针对GNN缺乏图像分类中常见的不变性学习技术的问题，提出缩放自我图概念，通过缩放边泛化传统自我图，构建ScaleNet架构，在多数据集上性能优于inception模型，为图学习提供统一框架。
摘要：
图神经网络（GNNs）推动了关系数据分析的发展，但缺乏图像分类中常见的不变性学习技术。在使用GNN进行节点分类时，实际被分类的是中心节点的自我图（ego-graph）。本研究通过类比图像处理，将尺度不变性概念扩展到节点分类：正如尺度不变性在图像分类中用于捕获多尺度特征，我们提出“缩放自我图”（scaled ego-graphs）的概念。缩放自我图通过用“缩放边”（scaled-edges）替换无向单边来泛化传统自我图，其中缩放边是多个有向边的有序序列。我们在七个基准数据集上实证评估了所提出的图中尺度不变性的性能，涵盖同构和异构结构。我们基于尺度不变性的图学习比源自随机游走的inception模型更简单、更快、更准确，性能更优。尺度不变性解释了inception模型在同构图上的成功及其在异构上的局限性。为确保inception模型也适用于异构结构，我们进一步提出ScaleNet架构，该架构利用多尺度特征。ScaleNet在七个数据集中的五个（四个同构、一个异构）上取得最先进结果，并在其余两个上匹配顶级性能，证明其出色的适用性。这代表了图学习的重大进展，提供了一个统一框架，可增强各种图类型的节点分类性能。我们的代码可在https://github.com/Qin87/ScaleNet/tree/July25获取。
英文摘要：
Graph Neural Networks (GNNs) have advanced relational data analysis but lack invariance learning techniques common in image classification. In node classification with GNNs, it is actually the ego-graph of the center node that is classified. This research extends the scale invariance concept to node classification by drawing an analogy to image processing: just as scale invariance being used in image classification to capture multi-scale features, we propose the concept of ``scaled ego-graphs''. Scaled ego-graphs generalize traditional ego-graphs by replacing undirected single-edges with ``scaled-edges'', which are ordered sequences of multiple directed edges. We empirically assess the performance of the proposed scale invariance in graphs on seven benchmark datasets, across both homophilic and heterophilic structures. Our scale-invariance-based graph learning outperforms inception models derived from random walks by being simpler, faster, and more accurate. The scale invariance explains inception models' success on homophilic graphs and limitations on heterophilic graphs. To ensure applicability of inception model to heterophilic graphs as well, we further present ScaleNet, an architecture that leverages multi-scaled features. ScaleNet achieves state-of-the-art results on five out of seven datasets (four homophilic and one heterophilic) and matches top performance on the remaining two, demonstrating its excellent applicability. This represents a significant advance in graph learning, offering a unified framework that enhances node classification across various graph types. Our code is available at https://github.com/Qin87/ScaleNet/tree/July25.
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。
