Coefficient Decomposition for Spectral Graph Convolution

arXiv:2405.03296v1 Announce Type: new
Abstract: Spectral graph convolutional network (SGCN) is a kind of graph neural networks (GNN) based on graph signal filters, and has shown compelling expressivity for modeling graph-structured data. Most SGCNs adopt polynomial filters and learn the coefficients from the training data. Many of them focus on which polynomial basis leads to optimal expressive power and models' architecture is little discussed. In this paper, we propose a general form in terms of spectral graph convolution, where the …

abstract arxiv convolution convolutional cs.ai cs.lg data filters focus gnn graph graph neural networks kind leads learn modeling network networks neural networks polynomial signal structured data them training training data type