16 April 2024 in IEEE Transactions on Computational Social Systems
Graph neural networks (GNNs) have achieved remarkable success in various real-world applications. However, recent studies highlight the vulnerability of GNNs to malicious perturbations. Previous adversaries primarily focus on graph modifications or node injections to existing graphs, yielding promising results but with notable limitations. Graph modification attack (GMA) requires manipulation of the original graph, which is often impractical, while graph injection attack (GIA) necessitates training a surrogate model in the black-box setting, leading to significant performance degradation due to divergence between the surrogate architecture and the actual victim model. Furthermore, most methods concentrate on a single attack goal and lack a generalizable adversary to develop distinct attack strategies for diverse goals, thus limiting precise control over victim model behavior in real-world scenarios. To address these issues, we present a gradient-free generalizable adversary that injects a single malicious node to manipulate the classification result of a target node in the black-box evasion setting. Specifically, we model the single-node injection label specificity attack as a Markov decision process (MDP) and propose gradient-free generalizable single node injection attack, namely G2-SNIA, a reinforcement learning framework employing proximal policy optimization (PPO). By directly querying the victim model, G2-SNIA learns patterns from exploration to achieve diverse attack goals with extremely limited attack budgets. Through comprehensive experiments over three acknowledged benchmark datasets and four prominent GNNs in the most challenging and realistic scenario, we demonstrate the superior performance of our proposed G2-SNIA over the existing state-of-the-art baselines. Moreover, by comparing G2-SNIA with multiple white-box evasion baselines, we confirm its capacity to generate solutions comparable to those of the best adversaries.
30 November 2023 in IEEE Transactions on Circuits and Systems II: Express Briefs
The clustering coefficient is a crucial characteristic of networks, providing insights into their local compactness. However, concerns arise regarding the clustering coefficient’s robustness when networks are subjected to adversarial attacks. In this paper, we introduce and formalize the concept of clustering coefficient attack and propose an efficient strategy to attack it by rewiring a small number of links. Our approach leverages Simulated Annealing (SA) to create a clustering coefficient attack method. We conduct extensive experiments on five real-world networks to validate the vulnerability of the clustering coefficient under adversarial attacks. Surprisingly, we find that disturbing as little as 1% of the edges can lead to an increase of over 50% in the clustering coefficient. These results underscore the pressing need to explore ways to enhance the clustering coefficient’s robustness against such attacks in future research. Addressing this challenge could be an intriguing avenue for further investigation.
2 April 2024 in Physica A: Statistical Mechanics and its Applications
22 January 2024 in IEEE Transactions on Circuits and Systems II: Express Briefs
The concept of k-core plays an important role in measuring the cohesiveness and engagement of a network. And recent studies have shown the vulnerability of k-core under adversarial attacks. However, there are few researchers concentrating on the vulnerability of individual nodes within k-core. Therefore, in this paper, we attempt to study Targeted k-Nodes Collapse Problem (TNsCP), which focuses on removing a minimal size set of edges to make multiple target k-nodes collapse. For this purpose, we first propose a novel algorithm named MOD for candidate reduction. Then we introduce an efficient strategy named MONA, based on MOD, to address TNsCP. Extensive experiments validate the effectiveness and scalability of MONA compared to several baselines. An open-source implementation is available at https://github.com/Yocenly/MONA.