Robust Graph Neural Networks via Unbiased Aggregation
This work addresses the vulnerability of GNNs to adversarial attacks, which is a critical issue for applications relying on graph data, though it appears incremental as it builds on existing robust GNN analysis.
The paper tackles the adversarial robustness problem in Graph Neural Networks (GNNs) by proposing a robust and unbiased graph signal estimator, which is implemented via an efficient algorithm and integrated into GNN layers with theoretical guarantees, resulting in strong robustness confirmed in comprehensive experiments.
The adversarial robustness of Graph Neural Networks (GNNs) has been questioned due to the false sense of security uncovered by strong adaptive attacks despite the existence of numerous defenses. In this work, we delve into the robustness analysis of representative robust GNNs and provide a unified robust estimation point of view to understand their robustness and limitations. Our novel analysis of estimation bias motivates the design of a robust and unbiased graph signal estimator. We then develop an efficient Quasi-Newton Iterative Reweighted Least Squares algorithm to solve the estimation problem, which is unfolded as robust unbiased aggregation layers in GNNs with theoretical guarantees. Our comprehensive experiments confirm the strong robustness of our proposed model under various scenarios, and the ablation study provides a deep understanding of its advantages. Our code is available at https://github.com/chris-hzc/RUNG.