We present InfoMotif, a new semi-supervised, motif-regularized, learning framework over graphs. We overcome two key limitations of message passing in popular graph neural networks (GNNs): localization (a k-layer GNN cannot utilize features outside the k-hop neighborhood of the labeled training nodes) and over-smoothed (structurally indistinguishable) representations. We propose the concept of attributed structural roles of nodes based on their occurrence in different network motifs, independent of network proximity. Two nodes share attributed structural roles if they participate in topologically similar motif instances over co-varying sets of attributes. Further, InfoMotif achieves architecture independence by regularizing the node representations of arbitrary GNNs via mutual information maximization. Our training curriculum dynamically prioritizes multiple motifs in the learning process without relying on distributional assumptions in the underlying graph or the learning task. We integrate three state-of-the-art GNNs in our framework, to show significant gains (3-10% accuracy) across six diverse, real-world datasets. We see stronger gains for nodes with sparse training labels and diverse attributes in local neighborhood structures.

中文翻译：

---

超越局部图神经网络：一个属性化的主题正则化框架

我们提出了 InfoMotif，这是一种新的半监督、主题正则化的图学习框架。我们克服了流行图神经网络 (GNN) 中消息传递的两个关键限制：定位（k 层 GNN 不能利用标记训练节点的 k 跳邻域之外的特征）和过度平滑（结构上无法区分）表示。我们提出了基于节点在不同网络基序中出现的属性结构角色的概念，与网络接近度无关。如果两个节点在共变属性集上参与拓扑相似的主题实例，则它们共享属性结构角色。此外，InfoMotif 通过互信息最大化来规范任意 GNN 的节点表示，从而实现架构独立。我们的培训课程在学习过程中动态地优先考虑多个主题，而不依赖于底层图形或学习任务中的分布假设。我们在我们的框架中集成了三个最先进的 GNN，以在六个不同的真实世界数据集上显示出显着的收益（3-10% 的准确率）。我们看到在局部邻域结构中具有稀疏训练标签和不同属性的节点有更强的收益。