Graph Convolutional Neural Networks via Scattering

TL;DR

This paper introduces a graph convolutional neural network based on scattering transforms, achieving permutation invariance and stability, with competitive results on benchmark datasets.

Contribution

It generalizes the scattering transform to graphs, creating a new convolutional neural network architecture with theoretical invariance and stability guarantees.

Findings

Achieves permutation invariance in graph features

Demonstrates stability to graph manipulations

Shows competitive performance on datasets

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 graph manipulations. Numerical results demonstrate competitive performance on relevant datasets.