Research map / Graph learning / GNN architectures
Principal Neighbourhood Aggregation for Graph Nets
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 |
|---|---|---|---|
| combine multiple neighborhood aggregators with degree-based scalers | GNN layers · Multi-task artificial benchmark | log10 MSE | Helped |
| use PNA rather than single-aggregator GNN layers | GNNs · Larger-graph extrapolation | MSE relative to baseline | Helped |
| remove degree scalers (ablation) | PNA · Chemical benchmarks ZINC and MolHIV | Dataset-specific | Hurt |
| remove degree scalers (ablation) | PNA · Computer vision benchmarks CIFAR10 and MNIST | Dataset-specific | Helped |
| add more moment aggregators (ablation) | PNA · Multi-task artificial benchmark | log10 MSE | Mixed |
| use max and min aggregators instead of higher moments (ablation) | PNA · Multi-task artificial benchmark | log10 MSE | Helped |
| replace max and min aggregators with softmax and softmin (ablation) | PNA · Benchmarks | Performance | Hurt |
Closest papers by meaning
- Fixed Aggregation Features Can Rival GNNs (ICML 2026 · similarity 0.90)
- Graph Neural Networks Are More Than Filters: Revisiting and Benchmarking from A Spectral Perspective (ICLR 2025 · similarity 0.90)
- Covered Forest: Fine-grained generalization analysis of graph neural networks (ICML 2025 · similarity 0.89)
- Sequential Signal Mixing Aggregation for Message Passing Graph Neural Networks (NeurIPS 2024 · similarity 0.88)
- Integrating Co-Training with Edge Discrimination to Enhance Graph Neural Networks Under Heterophily (AAAI 2025 · similarity 0.88)
- Making Classic GNNs Strong Baselines Across Varying Homophily: A Smoothness–Generalization Perspective (NeurIPS 2025 · similarity 0.87)
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.