图神经网络的泛化和表征限制

Feb, 2020

Generalization and Representational Limits of Graph Neural Networks

Vikas K. Garg, Stefanie Jegelka, Tommi Jaakkola

TL;DR本研究关注图神经网络的本质问题：无法仅仅依靠局部信息计算多个重要的图形特征；同时提出了信息传递图神经网络的第一个数据相关泛化界限，这一分析专门考虑了 GNN 局部置换不变性，该边界比现有的基于 VC 维度的 GNN 保证要紧密，与循环神经网络的 Rademacher 界限相当。

Abstract

We address two fundamental questions about graph neural networks (GNNs). First, we prove that several important graph properties cannot be computed by GNNs that rely entirely on local information. Such GNNs include the standard →