Diffusion Improves Graph Learning

NeurIPS 2019 Johannes KlicperaStefan WeißenbergerStephan Günnemann

Graph convolution is the core of most Graph Neural Networks (GNNs) and usually approximated by message passing between direct (one-hop) neighbors. In this work, we remove the restriction of using only the direct neighbors by introducing a powerful, yet spatially localized graph convolution: Graph diffusion convolution (GDC)... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Node Classification AMZ Comp GCN (Heat Diffusion) Accuracy 86.77% # 1
Node Classification AMZ Photo JK (Heat Diffusion) Accuracy 92.93% # 1
Node Classification Citeseer GCN (PPR Diffusion) Accuracy 73.35% # 10
Node Classification Coauthor CS GCN (PPR Diffusion) Accuracy 93.01% # 1
Node Classification Pubmed JK (Heat Diffusion) Accuracy 79.95% # 11