Simple and Deep Graph Convolutional Networks

TL;DR

This paper introduces GCNII, a deep graph convolutional network that addresses over-smoothing with residual and identity techniques, demonstrating superior performance on multiple graph learning tasks.

Contribution

The paper proposes GCNII, a novel deep GCN architecture with residual and identity mappings, effectively mitigating over-smoothing and improving performance.

Findings

GCNII outperforms existing methods on various tasks.

The techniques of residual and identity mappings effectively prevent over-smoothing.

Deep GCNs can be successfully trained with the proposed methods.

Abstract

Graph convolutional networks (GCNs) are a powerful deep learning approach for graph-structured data. Recently, GCNs and subsequent variants have shown superior performance in various application areas on real-world datasets. Despite their success, most of the current GCN models are shallow, due to the {\em over-smoothing} problem. In this paper, we study the problem of designing and analyzing deep graph convolutional networks. We propose the GCNII, an extension of the vanilla GCN model with two simple yet effective techniques: {\em Initial residual} and {\em Identity mapping}. We provide theoretical and empirical evidence that the two techniques effectively relieves the problem of over-smoothing. Our experiments show that the deep GCNII model outperforms the state-of-the-art methods on various semi- and full-supervised tasks. Code is available at https://github.com/chennnM/GCNII .