English(EN) PAC-Bayesian Adversarially Robust Generalization for Message Passing Graph Neural Networks: A Sensitivity Analysis

新的 PAC-Bayesian 框架增强了 GNN 的对抗鲁棒性分析

作者 PulseAugur 编辑部 · [2 个来源] · 2026-06-04 15:31

研究人员开发了一个新的 PAC-Bayesian 框架来分析消息传递图神经网络 (MPGNN) 的对抗鲁棒性。该框架通过量化参数敏感性并使用各向异性高斯后验来提供更紧密的泛化界限。该分析改进了谱范数依赖性并降低了复杂性因素，旨在指导 MPGNN 的设计以提高对抗鲁棒性。 AI

影响为设计更能抵御对抗性攻击的安全图神经网络提供了更精细的理论理解。

排序理由该集群包含一篇详细介绍图神经网络新分析框架的学术论文。

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv stat.ML TIER_1 English(EN) · Ziling Liang, Xinping Yi, Qingsong Wen, Shi Jin · 2026-06-05 04:00

PAC-贝叶斯对抗鲁棒性泛化用于消息传递图神经网络：一项敏感性分析

arXiv:2606.06293v1 Announce Type: cross Abstract: Whilst the vulnerability of graph neural networks (GNNs) to adversarial attacks poses a critical threat to graph representation learning, the understanding of the robust generalization behavior remains a fundamental challenge in t…
arXiv stat.ML TIER_1 English(EN) · Shi Jin · 2026-06-04 15:31

PAC-贝叶斯对抗鲁棒性泛化在消息传递图神经网络中的应用：一项敏感性分析

Whilst the vulnerability of graph neural networks (GNNs) to adversarial attacks poses a critical threat to graph representation learning, the understanding of the robust generalization behavior remains a fundamental challenge in the adversarial setting. Recently, PAC-Bayesian mar…