AI Paper Atlas

Research map / Graph learning / GNN architectures

Gradient Gating for Deep Multi-Rate Learning on Graphs

T. Konstantin Rusch, Benjamin Paul Chamberlain, Michael W. Mahoney, Michael M. Bronstein, Siddhartha Mishra

ICLR 2023 · Graph learning › GNN architectures · approach: Expressive equivariant message passing · 7 citations (OpenAlex)

Read the paper

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
Add gradient gating to GCN or GATGCN and GAT · Chameleon and Squirrel multi-scale node regressionnormalized test MSE: Chameleon: 0.207 ± 0.039 -> 0.137 ± 0.033 (GCN); 0.207 ± 0.038 -> 0.136 ± 0.029 (GAT). Squirrel: 0.143 ± 0.039 -> 0.070 Improved
Add gradient gating to message passingGCN and GAT · 1000-layer grid propagationDirichlet energyImproved
Use gradient gating in deeper GCNsGCN on Cora · Cora test classification, 2 to 128 layersaccuracyImproved
Add gradient gating to GCN or GATGCN and GAT · Synthetic Cora at strongest heterophilyaccuracy: over 80% versus approximately 20%Improved
Add gradient gating to GraphSAGEGraphSAGE · Texas, Wisconsin, Film, Squirrel, Chameleon, Cornellaccuracy: average improvement of 18.6%Mixed
Add gradient gating to GATGAT · Texas, Wisconsin, Film, Squirrel, Chameleon, Cornellaccuracy: average improvement of 45.75%Mixed
Add gradient gating to GCNGCN · Texas, Wisconsin, Film, Squirrel, Chameleon, Cornellaccuracy: average improvement of 45.4%Mixed
Use gradient-gated GraphSAGEGraphSAGE · snap-patents and arXiv-yearaccuracy: by up to 13%Improved
Increase the multi-rate strength α (ablation)G²-GCN and G²-GAT · Chameleon node regressiontest NMSEImproved
Tune p across values from 1 to 5 (ablation)G²-GraphSAGE · Texas, Squirrel, Film, Wisconsin, ChameleonaccuracyNo clear effect
Remove the additional rate-generating GNN (ablation)G²-GraphSAGE · Filmaccuracy: 37.14 ± 1.01 -> 36.83 ± 1.26Hurt
Remove the additional rate-generating GNN (ablation)G²-GraphSAGE · Squirrelaccuracy: 64.26 ± 2.38 -> 55.78 ± 1.61Hurt

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