Enhancing the Resilience of Graph Neural Networks to Topological Perturbations in Sparse Graphs
This work addresses robustness issues in GNNs for node classification, particularly in sparse graph scenarios, representing an incremental advancement over prior Bayesian approaches.
The paper tackles the vulnerability of graph neural networks (GNNs) to topological perturbations in sparse graphs by proposing TraTopo, a label inference framework that combines topology-driven propagation, Bayesian transitions, and random walks, resulting in significant accuracy improvements over existing methods.
Graph neural networks (GNNs) have been extensively employed in node classification. Nevertheless, recent studies indicate that GNNs are vulnerable to topological perturbations, such as adversarial attacks and edge disruptions. Considerable efforts have been devoted to mitigating these challenges. For example, pioneering Bayesian methodologies, including GraphSS and LlnDT, incorporate Bayesian label transitions and topology-based label sampling to strengthen the robustness of GNNs. However, GraphSS is hindered by slow convergence, while LlnDT faces challenges in sparse graphs. To overcome these limitations, we propose a novel label inference framework, TraTopo, which combines topology-driven label propagation, Bayesian label transitions, and link analysis via random walks. TraTopo significantly surpasses its predecessors on sparse graphs by utilizing random walk sampling, specifically targeting isolated nodes for link prediction, thus enhancing its effectiveness in topological sampling contexts. Additionally, TraTopo employs a shortest-path strategy to refine link prediction, thereby reducing predictive overhead and improving label inference accuracy. Empirical evaluations highlight TraTopo's superiority in node classification, significantly exceeding contemporary GCN models in accuracy.