Research map / Graph learning / GNN architectures
Limitless Stability for Graph Convolutional Networks
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 the unnormalized graph Laplacian for collapsing strongly connected subgraphs into single nodes | Graph convolutional networks · Strong-edge graph collapse | transferability error bound: K2√δ | Helped |
| Use adjacency matrices or normalized graph Laplacians instead of the unnormalized graph Laplacian | Graph convolutional networks · Strong-edge graph collapse | transferability stability | Hurt |
| Compare resolvents rather than graph Laplacians when measuring structural change | Graph convolutional networks · Strong-edge graph collapse | operator difference: O(δ_a·δ_b) | Helped |
| Increase the coupling strength of the collapsing subgraph | Graph convolutional networks · QM7 methane graph, effective versus original representation | graph-level transferability error: O(1) | Helped |
| Use p-norm aggregation with p ≥ 2 | Graph convolutional networks · Graph-level aggregation | input-signal stability bound | No clear effect |
Closest papers by meaning
- On the Stability of Graph Convolutional Neural Networks: A Probabilistic Perspective (NeurIPS 2025 · similarity 0.92)
- Unitary Convolutions for Learning on Graphs and Groups (NeurIPS 2024 · similarity 0.87)
- A Manifold Perspective on the Statistical Generalization of Graph Neural Networks (ICML 2025 · similarity 0.87)
- Graph neural networks and non-commuting operators (NeurIPS 2024 · similarity 0.87)
- Spectral Graph Sparsification Preserves Representation Geometry in Graph Neural Networks (NeurIPS 2026 · similarity 0.87)
- Higher-Order Graphon Neural Networks: Approximation and Cut Distance (ICLR 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.