RRLFSOR: An Efficient Self-Supervised Learning Strategy of Graph Convolutional Networks

TL;DR

This paper introduces RRLFSOR, a self-supervised learning strategy for GCNs that enhances performance and over-smoothing resistance by randomly removing links, validated on citation datasets with significant accuracy improvements.

Contribution

The paper proposes RRLFSOR, a novel data augmentation method for GCNs that improves self-learning ability and performance, especially in low-label scenarios.

Findings

Outperforms baseline models by up to 21.34% in accuracy.

Effective on multiple GCN models and datasets.

Enhances over-smoothing resistance in GCNs.

Abstract

Graph Convolutional Networks (GCNs) are widely used in many applications yet still need large amounts of labelled data for training. Besides, the adjacency matrix of GCNs is stable, which makes the data processing strategy cannot efficiently adjust the quantity of training data from the built graph structures.To further improve the performance and the self-learning ability of GCNs,in this paper, we propose an efficient self-supervised learning strategy of GCNs,named randomly removed links with a fixed step at one region (RRLFSOR).RRLFSOR can be regarded as a new data augmenter to improve over-smoothing.RRLFSOR is examined on two efficient and representative GCN models with three public citation network datasets-Cora,PubMed,and Citeseer.Experiments on transductive link prediction tasks show that our strategy outperforms the baseline models consistently by up to 21.34% in terms of accuracy on three benchmark datasets.