Graph Defense Diffusion Model

TL;DR

The paper introduces GDDM, a diffusion-based graph purification model that effectively defends against multiple adversarial attacks by iteratively removing noise and restoring graph structures, outperforming existing methods.

Contribution

GDDM is a novel, flexible diffusion model-based approach that enhances graph defense by modeling attack processes and improving scalability across datasets.

Findings

GDDM outperforms state-of-the-art defenses on three real-world datasets.

It effectively defends against targeted and non-targeted adversarial attacks.

The model demonstrates strong scalability and transferability across datasets.

Abstract

Graph Neural Networks (GNNs) are highly vulnerable to adversarial attacks, which can greatly degrade their performance. Existing graph purification methods attempt to address this issue by filtering attacked graphs. However, they struggle to defend effectively against multiple types of adversarial attacks (e.g., targeted attacks and non-targeted attacks) simultaneously due to limited flexibility. Additionally, these methods lack comprehensive modeling of graph data, relying heavily on heuristic prior knowledge. To overcome these challenges, we introduce the Graph Defense Diffusion Model (GDDM), a flexible purification method that leverages the denoising and modeling capabilities of diffusion models. The iterative nature of diffusion models aligns well with the stepwise process of adversarial attacks, making them particularly suitable for defense. By iteratively adding and removing noises (edges), GDDM effectively purifies attacked graphs, restoring their original structures and features. Our GDDM consists of two key components: (1) Graph Structure-Driven Refiner, which preserves the basic fidelity of the graph during the denoising process, and ensures that the generated graph remains consistent with the original scope; and (2) Node Feature-Constrained Regularizer, which removes residual impurities from the denoised graph, further enhancing the purification effect. By designing tailored denoising strategies to handle different types of adversarial attacks, we improve the GDDM's adaptability to various attack scenarios. Furthermore, GDDM demonstrates strong scalability, leveraging its structural properties to seamlessly transfer across similar datasets without retraining. Extensive experiments on three real-world datasets demonstrate that GDDM outperforms state-of-the-art methods in defending against various adversarial attacks, showcasing its robustness and effectiveness.