How-Powerful-of-GNN
HOW POWERFUL ARE GRAPH NEURAL NETWORKS?
This paper:
- characterize how expressive different GNN variants are in learning to represent and distinguish between different graph structures
- show that GNNs are at most as powerful as the WL test in distinguishing graph structures.
- identify graph structures that cannot be distinguished by popular GNN variants, such as GCN (Kipf & Welling, 2017) and GraphSAGE (Hamilton et al., 2017a)
- develop a simple neural architecture, Graph Isomorphism Network (GIN)