BriefGPT.xyz
Oct, 2024
关于一维图神经网络的VC维的说明
A note on the VC dimension of 1-dimensional GNNs
HTML
PDF
Noah Daniëls, Floris Geerts
TL;DR
本研究关注图神经网络（GNNs）的一维模型的泛化能力，从VC维度的角度探讨其表现。我们扩展了之前的研究，证明单参数的一维GNN在无界图中具有无限的VC维，这表明即使是最简单的GNN在泛化能力上也存在固有的局限性。
Abstract
Graph Neural Networks
(GNNs) have become an essential tool for analyzing graph-structured data, leveraging their ability to capture complex relational information. While the
Expressivity
of GNNs, particularly the
→