Graph Wavelet Neural Network
This work addresses efficiency and interpretability issues in graph neural networks for researchers and practitioners in machine learning, though it is incremental as it builds on existing spectral methods.
The authors tackled the computational cost and lack of interpretability in spectral graph convolutional neural networks by proposing a graph wavelet neural network (GWNN) that uses graph wavelet transform instead of Fourier transform, achieving significant performance improvements in semi-supervised classification on benchmark datasets like Cora, Citeseer, and Pubmed.
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.