提出新噪声模型与方法，让图神经网络更抗实际噪声干扰。
Training Robust Graph Neural Networks by Modeling Noise Dependencies
- 建模节点特征噪声与图结构、标签的依赖关系，更贴近真实场景。
- 在多种噪声下实验表现优于现有方法，尤其在复杂依赖噪声中提升显著。
- 适合研究鲁棒图学习、实际部署中噪声敏感问题的开发者参考。
现实应用中，图节点特征常受多种来源噪声影响，导致图神经网络性能严重下降。尽管已有多种增强鲁棒性的方法，但均基于噪声与图结构及节点标签独立的不切实际假设，限制了其适用性。为此，本文提出更真实的噪声场景——依赖感知的图噪声（DANG），其中节点特征噪声会通过链式依赖传播至图结构和节点标签。我们提出新型鲁棒图神经网络DA-GNN，利用变分推断捕捉DANG数据生成过程中的因果关系。此外，构建了模拟真实应用场景的基准数据集，支持更实用的鲁棒图学习研究。大量实验表明，DA-GNN在包括DANG和传统噪声模型在内的多种场景下均持续优于现有基线。代码已开源：https://github.com/yeonjun-in/torch-DA-GNN。
原文摘要 · Abstract (English)
In real-world applications, node features in graphs often contain noise from various sources, leading to significant performance degradation in GNNs. Although several methods have been developed to enhance robustness, they rely on the unrealistic assumption that noise in node features is independent of the graph structure and node labels, thereby limiting their applicability. To this end, we introduce a more realistic noise scenario, dependency-aware noise on graphs (DANG), where noise in node features create a chain of noise dependencies that propagates to the graph structure and node labels. We propose a novel robust GNN, DA-GNN, which captures the causal relationships among variables in the data generating process (DGP) of DANG using variational inference. In addition, we present new benchmark datasets that simulate DANG in real-world applications, enabling more practical research on robust GNNs. Extensive experiments demonstrate that DA-GNN consistently outperforms existing baselines across various noise scenarios, including both DANG and conventional noise models commonly considered in this field. Our code is available at https://github.com/yeonjun-in/torch-DA-GNN.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。