SFR-GNN: Simple and Fast Robust GNNs against Structural Attacks
This addresses the problem of computational inefficiency in defending GNNs against structural attacks for researchers and practitioners, though it is incremental as it builds on existing robust models.
The paper tackles the vulnerability of Graph Neural Networks (GNNs) to adversarial structural attacks by proposing SFR-GNN, a defense method that uses pre-training and contrastive learning to achieve a 24%--162% speedup over existing robust models while maintaining robustness for node classification.
Graph Neural Networks (GNNs) have demonstrated commendable performance for graph-structured data. Yet, GNNs are often vulnerable to adversarial structural attacks as embedding generation relies on graph topology. Existing efforts are dedicated to purifying the maliciously modified structure or applying adaptive aggregation, thereby enhancing the robustness against adversarial structural attacks. It is inevitable for a defender to consume heavy computational costs due to lacking prior knowledge about modified structures. To this end, we propose an efficient defense method, called Simple and Fast Robust Graph Neural Network (SFR-GNN), supported by mutual information theory. The SFR-GNN first pre-trains a GNN model using node attributes and then fine-tunes it over the modified graph in the manner of contrastive learning, which is free of purifying modified structures and adaptive aggregation, thus achieving great efficiency gains. Consequently, SFR-GNN exhibits a 24%--162% speedup compared to advanced robust models, demonstrating superior robustness for node classification tasks.