Disttack: Graph Adversarial Attacks Toward Distributed GNN Training

TL;DR

Disttack is a novel adversarial attack framework targeting distributed GNN training, effectively disrupting gradient synchronization by injecting attacks into a single node, leading to significant performance degradation.

Contribution

This work introduces the first adversarial attack method specifically designed for distributed GNN training, exploiting gradient update patterns to enhance attack efficiency and impact.

Findings

Disttack amplifies accuracy degradation by 2.75 times compared to previous methods.

It achieves a 17.33 times speedup in attack execution.

The attack remains unnoticeable while significantly disrupting training.

Abstract

Graph Neural Networks (GNNs) have emerged as potent models for graph learning. Distributing the training process across multiple computing nodes is the most promising solution to address the challenges of ever-growing real-world graphs. However, current adversarial attack methods on GNNs neglect the characteristics and applications of the distributed scenario, leading to suboptimal performance and inefficiency in attacking distributed GNN training. In this study, we introduce Disttack, the first framework of adversarial attacks for distributed GNN training that leverages the characteristics of frequent gradient updates in a distributed system. Specifically, Disttack corrupts distributed GNN training by injecting adversarial attacks into one single computing node. The attacked subgraphs are precisely perturbed to induce an abnormal gradient ascent in backpropagation, disrupting gradient synchronization between computing nodes and thus leading to a significant performance decline of the trained GNN. We evaluate Disttack on four large real-world graphs by attacking five widely adopted GNNs. Compared with the state-of-the-art attack method, experimental results demonstrate that Disttack amplifies the model accuracy degradation by 2.75$\times$ and achieves speedup by 17.33$\times$ on average while maintaining unnoticeability.