Batch Virtual Adversarial Training for Graph Convolutional Networks

TL;DR

This paper introduces batch virtual adversarial training (BVAT), a new regularization technique for graph convolutional networks that enhances their robustness and achieves state-of-the-art results in semi-supervised node classification tasks.

Contribution

The paper proposes two novel BVAT algorithms for GCNs that improve model smoothness and performance on graph-structured data, outperforming existing methods.

Findings

BVAT improves GCN robustness against local perturbations.

The methods achieve state-of-the-art accuracy on citation networks.

Experimental results validate the effectiveness of BVAT.

Abstract

We present batch virtual adversarial training (BVAT), a novel regularization method for graph convolutional networks (GCNs). BVAT addresses the shortcoming of GCNs that do not consider the smoothness of the model's output distribution against local perturbations around the input. We propose two algorithms, sample-based BVAT and optimization-based BVAT, which are suitable to promote the smoothness of the model for graph-structured data by either finding virtual adversarial perturbations for a subset of nodes far from each other or generating virtual adversarial perturbations for all nodes with an optimization process. Extensive experiments on three citation network datasets Cora, Citeseer and Pubmed and a knowledge graph dataset Nell validate the effectiveness of the proposed method, which establishes state-of-the-art results in the semi-supervised node classification tasks.