Over-Squashing in Graph Neural Networks: A Comprehensive survey

TL;DR

This survey comprehensively reviews the challenge of over-squashing in Graph Neural Networks, exploring its causes, effects, mitigation strategies, and related limitations to guide future research.

Contribution

It provides an extensive taxonomy of methods to address over-squashing, analyzing their trade-offs and effectiveness in GNNs.

Findings

Graph rewiring improves long-range information flow.

Normalization and spectral methods mitigate over-squashing.

Benchmark datasets are identified for evaluating mitigation strategies.

Abstract

Graph Neural Networks (GNNs) revolutionize machine learning for graph-structured data, effectively capturing complex relationships. They disseminate information through interconnected nodes, but long-range interactions face challenges known as "over-squashing". This survey delves into the challenge of over-squashing in Graph Neural Networks (GNNs), where long-range information dissemination is hindered, impacting tasks reliant on intricate long-distance interactions. It comprehensively explores the causes, consequences, and mitigation strategies for over-squashing. Various methodologies are reviewed, including graph rewiring, novel normalization, spectral analysis, and curvature-based strategies, with a focus on their trade-offs and effectiveness. The survey also discusses the interplay between over-squashing and other GNN limitations, such as over-smoothing, and provides a taxonomy of models designed to address these issues in node and graph-level tasks. Benchmark datasets for performance evaluation are also detailed, making this survey a valuable resource for researchers and practitioners in the GNN field.