PulseAugur
实时 11:15:27
English(EN) Operational Feature Fingerprints of Graph Datasets via a White-Box Signal-Subspace Probe

研究人员开发WG-SRC探测器以分析图神经网络行为

研究人员开发了WG-SRC,这是一种新颖的白盒探测器,用于分析和诊断图神经网络中使用的图数据集。该工具用固定的图信号字典替换了标准的报文传递机制,从而更清晰地理解节点是如何被分类的。WG-SRC的诊断功能将数据集的行为分解为原始特征、低通传播、高通差值和类别几何形状等组成部分,为进一步分析和数据集修改提供了见解。 AI

影响 提供了一种新的诊断工具,用于理解图数据集的特征并提高图神经网络的性能。

排序理由 这是一篇详细介绍分析图数据集新方法的学术论文。

在 arXiv cs.LG 阅读 →

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

研究人员开发WG-SRC探测器以分析图神经网络行为

报道来源 [2]

  1. arXiv cs.LG TIER_1 English(EN) · Yuchen Xiong, Swee Keong Yeap, Zhen Hong Ban ·

    Operational Feature Fingerprints of Graph Datasets via a White-Box Signal-Subspace Probe

    arXiv:2604.22676v1 Announce Type: new Abstract: Graph neural networks achieve strong node-classification accuracy, but their learned message passing entangles ego attributes, neighborhood smoothing, high-pass graph differences, class geometry, and classifier boundaries in an opaq…

  2. arXiv cs.LG TIER_1 English(EN) · Zhen Hong Ban ·

    Operational Feature Fingerprints of Graph Datasets via a White-Box Signal-Subspace Probe

    Graph neural networks achieve strong node-classification accuracy, but their learned message passing entangles ego attributes, neighborhood smoothing, high-pass graph differences, class geometry, and classifier boundaries in an opaque representation. This obscures why a node is c…