Research map / Graph learning / GNN architectures
Parameterized Explainer for Graph Neural Network
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use a shared parameterized explainer to explain instances collectively instead of independently optimizing each instance | PGExplainer · BA-Shapes | Explanation AUC: 0.925 -> 0.963±0.011 | Helped |
| Use a shared parameterized explainer to explain instances collectively instead of independently optimizing each instance | PGExplainer · BA-Community | Explanation AUC: 0.836 -> 0.945±0.019 | Helped |
| Use a shared parameterized explainer to explain instances collectively instead of independently optimizing each instance | PGExplainer · Tree-Cycles | Explanation AUC: 0.948 -> 0.987±0.007 | Helped |
| Use a shared parameterized explainer to explain instances collectively instead of independently optimizing each instance | PGExplainer · Tree-Grid | Explanation AUC: 0.875 -> 0.907±0.014 | Helped |
| Use a shared parameterized explainer to explain instances collectively instead of independently optimizing each instance | PGExplainer · BA-2motifs | Explanation AUC: 0.773 -> 0.926±0.021 | Helped |
| Use a shared parameterized explainer to explain instances collectively instead of independently optimizing each instance | PGExplainer · MUTAG | Explanation AUC: 0.783 -> 0.873±0.013 | Helped |
| Train the explainer on more instances rather than a single instance | PGExplainer · Inductive explanation test sets | AUC | Helped |
| Add the entropy constraint (ablation) | PGExplainer · BA-Community | AUC | Helped |
| Add size and entropy regularization (ablation) | PGExplainer · Synthetic datasets except BA-Community | AUC | No clear effect |
| Add a connectivity constraint (ablation) | PGExplainer · Noisy BA-Shapes | Explanation connectivity | Helped |
Closest papers by meaning
- GraphTrail: Translating GNN Predictions into Human-Interpretable Logical Rules (NeurIPS 2024 · similarity 0.91)
- EiG-Search: Generating Edge-Induced Subgraphs for GNN Explanation in Linear Time (ICML 2024 · similarity 0.91)
- Generating In-Distribution Proxy Graphs for Explaining Graph Neural Networks (ICML 2024 · similarity 0.91)
- Factorized Explainer for Graph Neural Networks (AAAI 2024 · similarity 0.91)
- Stratified GNN Explanations through Sufficient Expansion (AAAI 2024 · similarity 0.90)
- GnnXemplar: Exemplars to Explanations - Natural Language Rules for Global GNN Interpretability (NeurIPS 2025 · similarity 0.90)
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.