Comprehensive-LP-GNN

This repository consists some tentative models that incorporate label propagation to graph neural networks for graph representation learning in node, link or graph levels.

Contents

• gcn-lp-filter consists some tentative models that incorporate label propagation and graph filtering theory to graph neural networks for node classification.
• graph-classifier-dgl consists some tentative models that incorporate label propagation and sparse coding to graph neural networks for graph pooling and classification, using some examples of DGL.
• graph-classifier-vi consists some tentative models that incorporate label propagation and variational inference models (e.g. VGAE, Planar flow, Normalizing flow, IAF, etc.) to graph neural networks for graph pooling and classification, based on Graph U-Nets .