AI Paper Atlas

Research map / Graph learning / GNN architectures

On Over-Squashing in Message Passing Neural Networks: The Impact of Width, Depth, and Topology

Francesco Di Giovanni, Lorenzo Giusti, Federico Barbero, Giulia Luise, Pietro Lio, Michael M. Bronstein

ICML 2023 · Graph learning › GNN architectures · approach: Graph rewiring and scalable message passing · 29 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase hidden widthMPNNs · Graph transfer tasks: CrossedRing, Ring, CliquePathtask performanceImproved
increase hidden width and weight magnitudeMPNNs · Long-range information propagationgeneralizationHurt
increase depth to propagate information between distant nodesMPNNs · Long-range dependenciesinformation sensitivityNo clear effect
use greater depth in the deep regimeMPNNs · Deep MPNNsloss gradient magnitudeHurt
use a topology with lower commute time between task-relevant nodesMPNNs · Graph transfer taskstask difficultyImproved
rewire the graph to increase its Cheeger constantMPNNs · Graph rewiringJacobian obstructionImproved
add edges that reduce effective resistance between nodesMPNNs · Spatial graph rewiringinformation exchangeImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy