Research map / Graph learning / GNN architectures
Generalization and Representational Limits of Graph Neural Networks
Vikas Garg, Stefanie Jegelka, Tommi Jaakkola
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use only local information in message passing | LU-GNNs · Graph properties including girth, circumference, diameter, radius, cycles, and k-clique | ability to decide graph property | Hurt |
| Use consistent port numbering to order neighbor messages | CPNGNN · Graph discrimination | ability to distinguish graphs | Improved |
| Change the consistent port ordering | CPNGNN · Graph discrimination | ability to distinguish graphs | Mixed |
| Use local directional angles and distances | DimeNet · Graph discrimination | ability to decide graph properties | Improved |
| Use only DimeNet's local angles and distances | DimeNet · Girth, circumference, diameter, radius, and total number of cycles | ability to decide graph property | Hurt |
| Add port ordering and plane-based geometric features to directional messages | H-DCPN · Graph discrimination | ability to distinguish graphs | Improved |
| Bound GNN complexity through local computation trees and account for local permutation invariance | GNN generalization analysis · Binary graph classification | generalization-bound tightness | Improved |
Closest papers by meaning
- On the Rademacher Complexity of Graph Neural Networks: Unifying Expressivity and Geometry (NeurIPS 2026 · similarity 0.92)
- A Manifold Perspective on the Statistical Generalization of Graph Neural Networks (ICML 2025 · similarity 0.92)
- Covered Forest: Fine-grained generalization analysis of graph neural networks (ICML 2025 · similarity 0.91)
- Almost Surely Asymptotically Constant Graph Neural Networks (NeurIPS 2024 · similarity 0.90)
- Towards Bridging Generalization and Expressivity of Graph Neural Networks (ICLR 2025 · similarity 0.90)
- What do Graph Neural Networks learn? Insights from Tropical Geometry (NeurIPS 2024 · 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.
