AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Graph Neural Networks with Adaptive Readouts

David Buterez, Jon Paul Janet, Steven J Kiddle, Dino Oglic, Pietro Liò

NeurIPS 2022 · Graph learning › Graph self-supervised learning · approach: Heterophily and imbalance on graphs · 25 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
replace sum/mean/max readouts with neural readoutsGNN · regression datasets (R2)R2 ratio: 36 of 45 configurations improvedImproved
replace sum/mean/max readouts with neural readoutsGNN · classification datasets (MCC)MCC: 93 of 147 improved; min ratio 0.79Mixed
increase set transformer heads from 1 to 8 (ablation)ST minimal GCN · QM9R2: 0.20 -> 0.35Improved
increase heads beyond 4 or add self-attention block (ablation)Set Transformer readout · QM9R2: 0.35 -> 0.38 (complex GCN, 4 -> 8 heads)No clear effect
use MLP readout instead of standard readouts on 1M+ graphsVGAE GCN · proprietary datasetsR2: 0.33 -> 0.78Improved
use neural readouts on 1.5M graph datasetVGAE GCN · proprietaryR2: 0.07 -> 0.64Improved
use neural readouts on 2M graph datasetVGAE GCN · proprietaryR2: 0.06 -> 0.52Improved
add more GNN layers (2,3,4)GNN/VGAE · proprietary 1M+ datasetstrain lossNo clear effect
increase parameter count by switching from MLP to STreadouts · QM9performanceNo clear effect
use neural readouts for convergenceGNN · QM9epochs to converge: 500-1000 -> under 100 epochsImproved
train MLP/GRU readouts and evaluate on node permutationsMLP/GRU readouts · QM9 permutationserrorHurt
use Janossy readouts instead of standardJanossy readout · QM9MAE and R2Improved

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