Research map / Graph learning / GNN architectures
On the equivalence between graph isomorphism testing and function approximation with GNNs
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 |
|---|---|---|---|
| add matrix-multiplication (ring) branch to order-2 G-invariant network | Order-2 Graph G-inv. · CSL (classification) | max accuracy (%): 10 -> 80 | Helped |
| add matrix-multiplication (ring) branch to order-2 G-invariant network | Order-2 Graph G-inv. · IMDBB | mean accuracy (%): 71.3 -> 73.0 | Helped |
| add matrix-multiplication (ring) branch to order-2 G-invariant network | Order-2 Graph G-inv. · IMDBM | mean accuracy (%): 48.6 -> 48.2 | Hurt |
| add matrix-multiplication (ring) branch to order-2 G-invariant network | Order-2 Graph G-inv. · PTC | accuracy (%): 59.5 -> 65.7 | Helped |
| add matrix-multiplication (ring) branch to order-2 G-invariant network | Order-2 Graph G-inv. · MUTAG | accuracy (%): 84.6 -> 86.8 | Helped |
| add matrix-multiplication (ring) branch to order-2 G-invariant network | Order-2 Graph G-inv. · COLLAB | accuracy (%): 77.9 -> 80.1 | Helped |
| add matrix-multiplication (ring) branch to order-2 G-invariant network | Order-2 Graph G-inv. · PROTEINS | accuracy (%): 75.2 -> 75.7 | No clear effect |
| use Ring-GNN instead of GIN | Ring-GNN · CSL (classification) | max accuracy (%): 10 -> 80 | Helped |
| add node degree as input feature to Ring-GNN (ablation) | Ring-GNN · IMDBM | mean accuracy (%): 48.2 -> 51.3 | Helped |
| add node degree as input feature to Ring-GNN (ablation) | Ring-GNN · IMDBB | mean accuracy (%): 73.0 -> 73.3 | No clear effect |
Closest papers by meaning
- On Graph Neural Networks versus Graph-Augmented MLPs (ICLR 2021 · similarity 0.90)
- What do Graph Neural Networks learn? Insights from Tropical Geometry (NeurIPS 2024 · similarity 0.88)
- Graph neural networks and non-commuting operators (NeurIPS 2024 · similarity 0.88)
- Covered Forest: Fine-grained generalization analysis of graph neural networks (ICML 2025 · similarity 0.87)
- Homomorphism Expressivity of Spectral Invariant Graph Neural Networks (ICLR 2025 · similarity 0.87)
- Towards Robust Fidelity for Evaluating Explainability of Graph Neural Networks (ICLR 2024 · 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.