Graph convolutional neural networks via scattering

We generalize the scattering transform to graphs and consequently construct a convolutional neural network on graphs. We show that under certain conditions, any feature generated by such a network is approximately invariant to permutations and stable to signal and graph manipulations. Numerical resu...

Full description

Saved in:
Bibliographic Details
Published inApplied and computational harmonic analysis Vol. 49; no. 3; pp. 1046 - 1074
Main Authors Zou, Dongmian, Lerman, Gilad
Format Journal Article
LanguageEnglish
Published Elsevier Inc 01.11.2020
Subjects
Feature learning
Graph convolution
Graph neural networks
Permutation invariance
Scattering transform
Spectral graph theory
Wavelets
Wavelets
62M45
42C40
Permutation invariance
97K30
Feature learning
Graph neural networks
Graph convolution
68T30
Scattering transform
Spectral graph theory
Online AccessGet full text

Cover

More Information
Summary:We generalize the scattering transform to graphs and consequently construct a convolutional neural network on graphs. We show that under certain conditions, any feature generated by such a network is approximately invariant to permutations and stable to signal and graph manipulations. Numerical results demonstrate competitive performance on relevant datasets.
ISSN:1063-5203
1096-603X
DOI:10.1016/j.acha.2019.06.003