A Robust graph attention network with dynamic adjusted Graph

TL;DR

This paper introduces RoGAT, a robust graph attention network that enhances resistance to adversarial attacks by dynamically adjusting edge weights and features, significantly improving stability on citation datasets.

Contribution

The paper proposes RoGAT, a novel GAT variant with dynamic attention scores and revised edge weights to improve robustness against adversarial perturbations.

Findings

RoGAT outperforms recent defensive methods against targeted attacks.

RoGAT effectively reduces the impact of adversarial noise on graph data.

Experimental results demonstrate improved robustness on citation datasets.

Abstract

Graph Attention Networks(GATs) are useful deep learning models to deal with the graph data. However, recent works show that the classical GAT is vulnerable to adversarial attacks. It degrades dramatically with slight perturbations. Therefore, how to enhance the robustness of GAT is a critical problem. Robust GAT(RoGAT) is proposed in this paper to improve the robustness of GAT based on the revision of the attention mechanism. Different from the original GAT, which uses the attention mechanism for different edges but is still sensitive to the perturbation, RoGAT adds an extra dynamic attention score progressively and improves the robustness. Firstly, RoGAT revises the edges weight based on the smoothness assumption which is quite common for ordinary graphs. Secondly, RoGAT further revises the features to suppress features' noise. Then, an extra attention score is generated by the dynamic edge's weight and can be used to reduce the impact of adversarial attacks. Different experiments against targeted and untargeted attacks on citation data on citation data demonstrate that RoGAT outperforms most of the recent defensive methods.