从实用视角重新审视图神经网络与距离编码
Revisiting graph neural networks and distance encoding from a practical view
AI总结：
从实用角度重新审视图神经网络与距离编码，解释 DE 如何提升链接预测与节点分类性能，通过实验验证其有效性并给出使用建议。
摘要：
图神经网络（GNNs）被广泛应用于基于图结构数据的应用中，如节点分类和链接预测。然而，GNNs 常常被当作黑盒工具使用，很少被深入研究它们是否适合可能具有各种属性的特定应用。最近提出的距离编码（DE）技术（Li 等人，2020）神奇地使 GNNs 在许多应用中表现良好，包括节点分类和链接预测。（Li 等人，2020）中提供的理论通过证明 DE 提高了 GNNs 的表示能力来支持 DE。然而，该理论如何相应地辅助应用并不明显。在此，我们从更实用的视角重新审视 GNNs 和 DE。我们想解释 DE 如何使 GNNs 适用于节点分类和链接预测。具体来说，对于链接预测，DE 可以被视为建立一对节点表示之间相关性的一种方式。对于节点分类，问题变得更加复杂，因为不同的分类任务可能持有表示不同物理含义的节点标签。我们关注最广泛考虑的节点分类场景，并将节点标签分为两类：社区类型和结构类型，然后分析 GNNs 采用的不同机制来预测这两类标签。我们还进行了大量实验，比较了八种不同的 GNNs 与 DE 配对配置，在八个真实世界图上预测节点标签。结果证明了 DE 在预测结构类型标签方面的统一有效性。最后，我们得出了关于如何在节点分类任务中正确使用 GNNs 和 DE 的三点结论。
英文摘要：
Graph neural networks (GNNs) are widely used in the applications based on graph structured data, such as node classification and link prediction. However, GNNs are often used as a black-box tool and rarely get in-depth investigated regarding whether they fit certain applications that may have various properties. A recently proposed technique distance encoding (DE) (Li et al. 2020) magically makes GNNs work well in many applications, including node classification and link prediction. The theory provided in (Li et al. 2020) supports DE by proving that DE improves the representation power of GNNs. However, it is not obvious how the theory assists the applications accordingly. Here, we revisit GNNs and DE from a more practical point of view. We want to explain how DE makes GNNs fit for node classification and link prediction. Specifically, for link prediction, DE can be viewed as a way to establish correlations between a pair of node representations. For node classification, the problem becomes more complicated as different classification tasks may hold node labels that indicate different physical meanings. We focus on the most widely-considered node classification scenarios and categorize the node labels into two types, community type and structure type, and then analyze different mechanisms that GNNs adopt to predict these two types of labels. We also run extensive experiments to compare eight different configurations of GNNs paired with DE to predict node labels over eight real-world graphs. The results demonstrate the uniform effectiveness of DE to predict structure-type labels. Lastly, we reach three pieces of conclusions on how to use GNNs and DE properly in tasks of node classification.
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。
