A Simple Spectral Failure Mode for Graph Convolutional Networks

Add code
Oct 27, 2020
Carey E. Priebe, Cencheng Shen, Ningyuan Huang, Tianyi Chen
Figure 1 for A Simple Spectral Failure Mode for Graph Convolutional Networks
Figure 2 for A Simple Spectral Failure Mode for Graph Convolutional Networks
Figure 3 for A Simple Spectral Failure Mode for Graph Convolutional Networks
Figure 4 for A Simple Spectral Failure Mode for Graph Convolutional Networks

Share this with someone who'll enjoy it:

Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon

We present a simple generative model in which spectral graph embedding for subsequent inference succeeds whereas unsupervised graph convolutional networks (GCN) fail. The geometrical insight is that the GCN is unable to look beyond the first non-informative spectral dimension.

View paper onarxiv icon

Share this with someone who'll enjoy it:

Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon