Graph Wavelet Neural Network

TL;DR

This paper introduces a graph wavelet neural network (GWNN) that uses graph wavelet transforms for efficient, interpretable graph convolution, outperforming previous spectral methods in semi-supervised classification tasks.

Contribution

The paper proposes GWNN, a novel graph CNN leveraging fast, localized wavelet transforms that avoid costly eigendecomposition, improving efficiency and interpretability.

Findings

GWNN outperforms previous spectral graph CNNs on benchmark datasets

Wavelet transforms enable sparse, localized, and efficient graph convolutions

The method achieves superior semi-supervised classification accuracy

Abstract

We present graph wavelet neural network (GWNN), a novel graph convolutional neural network (CNN), leveraging graph wavelet transform to address the shortcomings of previous spectral graph CNN methods that depend on graph Fourier transform. Different from graph Fourier transform, graph wavelet transform can be obtained via a fast algorithm without requiring matrix eigendecomposition with high computational cost. Moreover, graph wavelets are sparse and localized in vertex domain, offering high efficiency and good interpretability for graph convolution. The proposed GWNN significantly outperforms previous spectral graph CNNs in the task of graph-based semi-supervised classification on three benchmark datasets: Cora, Citeseer and Pubmed.