Graph convolutional neural networks via scattering

Dongmian Zou; Gilad Lerman

doi:10.1016/j.acha.2019.06.003

Graph convolutional neural networks via scattering

Dongmian Zou, Gilad Lerman

Mathematics

Research output: Contribution to journal › Article › peer-review

52 Scopus citations

Abstract

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.

Original language	English (US)
Pages (from-to)	1046-1074
Number of pages	29
Journal	Applied and Computational Harmonic Analysis
Volume	49
Issue number	3
DOIs	https://doi.org/10.1016/j.acha.2019.06.003
State	Published - Nov 2020

Bibliographical note

Publisher Copyright:
© 2019 Elsevier Inc.

Keywords

Feature learning
Graph convolution
Graph neural networks
Permutation invariance
Scattering transform
Spectral graph theory
Wavelets

Access

10.1016/j.acha.2019.06.003

OpenUrl availability

Full text

Cite this

@article{d175e6d2a3864d278a9943420f314e48,

title = "Graph convolutional neural networks via scattering",

abstract = "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.",

keywords = "Feature learning, Graph convolution, Graph neural networks, Permutation invariance, Scattering transform, Spectral graph theory, Wavelets",

author = "Dongmian Zou and Gilad Lerman",

note = "Publisher Copyright: {\textcopyright} 2019 Elsevier Inc.",

year = "2020",

month = nov,

doi = "10.1016/j.acha.2019.06.003",

language = "English (US)",

volume = "49",

pages = "1046--1074",

journal = "Applied and Computational Harmonic Analysis",

issn = "1063-5203",

publisher = "Academic Press Inc.",

number = "3",

}

TY - JOUR

T1 - Graph convolutional neural networks via scattering

AU - Zou, Dongmian

AU - Lerman, Gilad

PY - 2020/11

Y1 - 2020/11

N2 - 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.

AB - 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.

KW - Feature learning

KW - Graph convolution

KW - Graph neural networks

KW - Permutation invariance

KW - Scattering transform

KW - Spectral graph theory

KW - Wavelets

UR - http://www.scopus.com/inward/record.url?scp=85067294243&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067294243&partnerID=8YFLogxK

U2 - 10.1016/j.acha.2019.06.003

DO - 10.1016/j.acha.2019.06.003

M3 - Article

AN - SCOPUS:85067294243

SN - 1063-5203

VL - 49

SP - 1046

EP - 1074

JO - Applied and Computational Harmonic Analysis

JF - Applied and Computational Harmonic Analysis

IS - 3

ER -

Graph convolutional neural networks via scattering

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this