AI Paper Atlas

Research map / Graph learning / GNN architectures

Train Once and Explain Everywhere: Pre-training Interpretable Graph Neural Networks

Jun Yin, Chaozhuo Li, Hao Yan, Jianxun Lian, Senzhang Wang

NeurIPS 2023 · Graph learning › GNN architectures · approach: Graph structure learning · 0 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
add explainer pre-training on synthetic PT-Motifs (ablation)π-GNN · Mutagexplanation ROC-AUC: 95.29 -> 99.81Improved
add explainer pre-training on synthetic PT-Motifs (ablation)π-GNN · BA-2Motifsexplanation ROC-AUC: 93.19 -> 99.33Improved
add explainer pre-training on synthetic PT-Motifs (ablation)π-GNN · Spurious-Motif b=0.9explanation ROC-AUC: 89.64 -> 96.39Improved
add explainer pre-training on synthetic PT-Motifs (ablation)π-GNN · Graph-SST2accuracy: 83.48 -> 88.05Improved
add explainer pre-training on synthetic PT-Motifs (ablation)π-GNN · Spurious-Motif b=0.7accuracy: 71.02 -> 77.52Improved
add structural pattern learning module (vs N-thread GNN encoders) (ablation)π-GNN · Mutag interpretation / Graph-SST2 predictionROC-AUC / accuracy: 3.85% / 1.08% average decrease when removedImproved
add hypergraph refining module (ablation)π-GNN · Mutag interpretation / Graph-SST2 predictionROC-AUC / accuracy: 6.56% / 1.84% average decrease when removedImproved
keep both modules (vs removing both, π-GNN-ALL) (ablation)π-GNN · Mutag interpretation / Graph-SST2 predictionROC-AUC / accuracy: 7.56% / 2.16% average decrease when removedImproved
shrink pre-training set from 80,000 to 10,000 graphs (PT-Motifs-S) (ablation)π-GNN · Mutagexplanation ROC-AUC: 99.81 -> 96.28Hurt
shrink pre-training set from 80,000 to 50,000 graphs (PT-Motifs-M) (ablation)π-GNN · Graph-SST2accuracy: 88.05 -> 87.59No clear effect
pre-train on small PT-Motifs-S instead of no pre-training (ablation)π-GNN · Ogbg-MolhivAUC: 79.71 -> 79.82No clear effect
pre-train on heavily imbalanced dataset (b = 0.7, i about 9.3) (ablation)π-GNN · Mutag, Molhiv, Graph-SST2interpretation/predictionHurt

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