English

Revisiting Graph Convolutional Network on Semi-Supervised Node Classification from an Optimization Perspective

Machine Learning 2020-09-28 v2 Artificial Intelligence Machine Learning

Abstract

Graph convolutional networks (GCNs) have achieved promising performance on various graph-based tasks. However they suffer from over-smoothing when stacking more layers. In this paper, we present a quantitative study on this observation and develop novel insights towards the deeper GCN. First, we interpret the current graph convolutional operations from an optimization perspective and argue that over-smoothing is mainly caused by the naive first-order approximation of the solution to the optimization problem. Subsequently, we introduce two metrics to measure the over-smoothing on node-level tasks. Specifically, we calculate the fraction of the pairwise distance between connected and disconnected nodes to the overall distance respectively. Based on our theoretical and empirical analysis, we establish a universal theoretical framework of GCN from an optimization perspective and derive a novel convolutional kernel named GCN+ which has lower parameter amount while relieving the over-smoothing inherently. Extensive experiments on real-world datasets demonstrate the superior performance of GCN+ over state-of-the-art baseline methods on the node classification tasks.

Keywords

Cite

@article{arxiv.2009.11469,
  title  = {Revisiting Graph Convolutional Network on Semi-Supervised Node Classification from an Optimization Perspective},
  author = {Hongwei Zhang and Tijin Yan and Zenjun Xie and Yuanqing Xia and Yuan Zhang},
  journal= {arXiv preprint arXiv:2009.11469},
  year   = {2020}
}
R2 v1 2026-06-23T18:45:30.165Z