When Does Self-Supervision Help Graph Convolutional Networks?

TL;DR

This paper systematically explores the integration of self-supervision into graph convolutional networks (GCNs), proposing new tasks and mechanisms that enhance their generalizability and robustness.

Contribution

It introduces the first comprehensive assessment of self-supervision in GCNs, proposing novel tasks and mechanisms, and analyzing their impact on GCN performance.

Findings

Self-supervision improves GCN generalizability.

Proper task design enhances robustness.

Multi-task self-supervision benefits adversarial training.

Abstract

Self-supervision as an emerging technique has been employed to train convolutional neural networks (CNNs) for more transferrable, generalizable, and robust representation learning of images. Its introduction to graph convolutional networks (GCNs) operating on graph data is however rarely explored. In this study, we report the first systematic exploration and assessment of incorporating self-supervision into GCNs. We first elaborate three mechanisms to incorporate self-supervision into GCNs, analyze the limitations of pretraining & finetuning and self-training, and proceed to focus on multi-task learning. Moreover, we propose to investigate three novel self-supervised learning tasks for GCNs with theoretical rationales and numerical comparisons. Lastly, we further integrate multi-task self-supervision into graph adversarial training. Our results show that, with properly designed task forms and incorporation mechanisms, self-supervision benefits GCNs in gaining more generalizability and robustness. Our codes are available at https://github.com/Shen-Lab/SS-GCNs.