TL;DR本研究分析了GCN、GCN with bias、ResGCN和APPNP模型的节点特征收敛过程,并提出了DropEdge来缓解过平滑问题,其在模拟数据和多个真实基准测试上均表现出显著性能提升。
Abstract
\emph{Over-fitting} and \emph{over-smoothing} are two main obstacles of developing deep Graph Convolutional Networks (GCNs) for node classification. In particular, over-fitting weakens the generalization ability