LGMLMar 9, 2020

A Survey on The Expressive Power of Graph Neural Networks

arXiv:2003.04078v4196 citations
AI Analysis

It synthesizes existing research on GNN expressiveness for researchers in machine learning and graph theory, but is incremental as it is a survey.

This survey addresses the theoretical limitations of graph neural networks (GNNs) by providing a comprehensive overview of their expressive power and provably powerful variants, without presenting new experimental results.

Graph neural networks (GNNs) are effective machine learning models for various graph learning problems. Despite their empirical successes, the theoretical limitations of GNNs have been revealed recently. Consequently, many GNN models have been proposed to overcome these limitations. In this survey, we provide a comprehensive overview of the expressive power of GNNs and provably powerful variants of GNNs.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes