arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.04038stat.MLcs.LG
收藏

图神经网络的PAC-贝叶斯对抗鲁棒泛化界

PAC-Bayesian Adversarially Robust Generalization Bounds for Graph Neural Network

Center for Data Science, Zhejiang University(浙江大学数据科学中心)

Tan Sun, Junhong Lin

AI总结:

本文利用PAC-贝叶斯框架为GCN和消息传递图神经网络建立对抗鲁棒泛化界,揭示扩散矩阵谱范数、权重谱范数和扰动因子的关键作用,并避免了对最大节点度的指数依赖,改进了现有结果。

摘要:

图神经网络(GNNs)已在各类图相关任务中获得广泛应用。然而,与深度神经网络类似,GNNs同样容易受到对抗攻击。实证研究表明,对抗鲁棒泛化在建立有效的对抗攻击防御算法中具有关键作用。本文利用PAC-贝叶斯框架,为两类流行的GNN——图卷积网络(GCN)和消息传递图神经网络——提供了对抗鲁棒泛化界。我们的结果揭示,图上扩散矩阵的谱范数、权重的谱范数以及扰动因子共同决定了两种模型的鲁棒泛化界。我们的界是将(Liao等,2020)中发展的结果从标准设置到对抗设置的非平凡推广,同时避免了对最大节点度的指数依赖。作为推论,我们在标准设置下为GCN推导出更好的PAC-贝叶斯鲁棒泛化界,通过避免对最大节点度的指数依赖,改进了(Liao等,2020)中的界。

英文摘要:

Graph neural networks (GNNs) have gained popularity for various graph-related tasks. However, similar to deep neural networks, GNNs are also vulnerable to adversarial attacks. Empirical studies have shown that adversarially robust generalization has a pivotal role in establishing effective defense algorithms against adversarial attacks. In this paper, we contribute by providing adversarially robust generalization bounds for two kinds of popular GNNs, graph convolutional network (GCN) and message passing graph neural network, using the PAC-Bayesian framework. Our result reveals that spectral norm of the diffusion matrix on the graph and spectral norm of the weights as well as the perturbation factor govern the robust generalization bounds of both models. Our bounds are nontrivial generalizations of the results developed in (Liao et al., 2020) from the standard setting to adversarial setting while avoiding exponential dependence of the maximum node degree. As corollaries, we derive better PAC-Bayesian robust generalization bounds for GCN in the standard setting, which improve the bounds in (Liao et al., 2020) by avoiding exponential dependence on the maximum node degree.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑