Robust Graph Data Learning via Latent Graph Convolutional Representation

TL;DR

This paper introduces LatGCR, a novel method for robust graph data representation that learns a flexible latent graph to improve capacity and resist structural noise and attacks, applicable in both supervised and unsupervised settings.

Contribution

The paper proposes LatGCR, a new latent graph convolutional approach that enhances representation capacity and robustness against structural perturbations in graph learning.

Findings

LatGCR outperforms existing methods on multiple datasets.

LatGCR demonstrates robustness to structural attacks and noise.

LatGCR is effective in both supervised and unsupervised tasks.

Abstract

Graph Convolutional Representation (GCR) has achieved impressive performance for graph data representation. However, existing GCR is generally defined on the input fixed graph which may restrict the representation capacity and also be vulnerable to the structural attacks and noises. To address this issue, we propose a novel Latent Graph Convolutional Representation (LatGCR) for robust graph data representation and learning. Our LatGCR is derived based on reformulating graph convolutional representation from the aspect of graph neighborhood reconstruction. Given an input graph $\textbf{A}$, LatGCR aims to generate a flexible latent graph $\widetilde{\textbf{A}}$ for graph convolutional representation which obviously enhances the representation capacity and also performs robustly w.r.t graph structural attacks and noises. Moreover, LatGCR is implemented in a self-supervised manner and thus provides a basic block for both supervised and unsupervised graph learning tasks. Experiments on several datasets demonstrate the effectiveness and robustness of LatGCR.