How powerful are Graph Convolutions? (review of Kipf & Welling, 2016)
This post is about a paper that has just come out recently on practical generalizations of convolutional layers to graphs: Thomas N. Kipf and Max Welling (2016) Semi-Supervised Classification with Graph Convolutional Networks Along the way I found this earlier, related paper: Defferrard, Bresson and Vandergheynst (NIPS 2016) Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering This post is mainly a review of (Kipf and Welling, 2016). The paper is nice to read, and while I like the general idea, I feel like the approximations made in the paper are too limiting and severely hurt the generality of the models we can build. This post explains why.
Relations
gives cons of Graph convolutional networks (GCN)
Generalization of neural networks to arbitrary graphs.
reviews Graph convolutional networks
Many important real-world datasets come in the form of graphs or networks: social networks, knowledge...
Edit details Edit relations Attach new author Attach new topic Attach new resource
from 1 review
- Resource level 4.0 /10
- beginner intermediate advanced
- Resource clarity 9.0 /10
- hardly clear sometimes unclear perfectly clear
- Reviewer's background 3.0 /10
- none basics intermediate advanced expert