Leveraging Vulnerabilities in Temporal Graph Neural Networks via Strategic High-Impact Assaults

2025-09-29

AI Summary

Existing adversarial attacks against Temporal Graph Neural Networks (TGNNs) suffer from simplistic perturbation strategies, poor stealth, and limited efficacy. To address these limitations, this paper proposes the High-Impact Attack (HIA) framework—a data-driven black-box attack that jointly assesses node importance across both structural and temporal dimensions. HIA constructs a surrogate model to identify critical nodes in these dual dimensions and employs a hybrid perturbation strategy combining edge injection with targeted edge deletion. This enables highly stealthy attacks under extremely low perturbation budgets. Extensive experiments on four state-of-the-art TGNNs—TGN, JODIE, DySAT, and TGAT—across five real-world temporal graph datasets demonstrate that HIA reduces Mean Reciprocal Rank (MRR) in link prediction by up to 35.55%, significantly outperforming prior methods. Crucially, HIA is the first work to systematically expose the deep vulnerability of TGNNs to coordinated structural-temporal adversarial perturbations.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Computer Vision: Adversarial Attacks & Robustness
• Planning, Routing, and Scheduling: Temporal Planning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Temporal Graph Neural Networks (TGNNs) have become indispensable for analyzing dynamic graphs in critical applications such as social networks, communication systems, and financial networks. However, the robustness of TGNNs against adversarial attacks, particularly sophisticated attacks that exploit the temporal dimension, remains a significant challenge. Existing attack methods for Spatio-Temporal Dynamic Graphs (STDGs) often rely on simplistic, easily detectable perturbations (e.g., random edge additions/deletions) and fail to strategically target the most influential nodes and edges for maximum impact. We introduce the High Impact Attack (HIA), a novel restricted black-box attack framework specifically designed to overcome these limitations and expose critical vulnerabilities in TGNNs. HIA leverages a data-driven surrogate model to identify structurally important nodes (central to network connectivity) and dynamically important nodes (critical for the graph's temporal evolution). It then employs a hybrid perturbation strategy, combining strategic edge injection (to create misleading connections) and targeted edge deletion (to disrupt essential pathways), maximizing TGNN performance degradation. Importantly, HIA minimizes the number of perturbations to enhance stealth, making it more challenging to detect. Comprehensive experiments on five real-world datasets and four representative TGNN architectures (TGN, JODIE, DySAT, and TGAT) demonstrate that HIA significantly reduces TGNN accuracy on the link prediction task, achieving up to a 35.55% decrease in Mean Reciprocal Rank (MRR) - a substantial improvement over state-of-the-art baselines. These results highlight fundamental vulnerabilities in current STDG models and underscore the urgent need for robust defenses that account for both structural and temporal dynamics.

Problem

• Exposing vulnerabilities in Temporal Graph Neural Networks via adversarial attacks
• Overcoming simplistic perturbations by targeting influential nodes and edges
• Developing stealthy hybrid attacks combining edge injection and deletion

Innovation

• HIA uses data-driven surrogate model to identify key nodes
• Hybrid perturbation combines edge injection and deletion
• Minimizes perturbations for stealth and maximizes performance degradation