Research map / Graph learning / GNN architectures
Deeper Insights Into Deep Graph Convolutional Networks: Stability and Generalization
Guangrui Yang, Ming Li, Feng Han, Xiaosheng Zhuang
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 normalized graph filter instead of random walk filter (ablation) | 5-layer GCN · Cora | generalization gap: 1.796 -> 1.352 | Improved |
| use normalized graph filter instead of random walk filter (ablation) | 5-layer GCN · Citeseer | generalization gap: 3.061 -> 2.661 | Improved |
| use normalized graph filter instead of random walk filter (ablation) | 5-layer GCN · Pubmed | generalization gap: 2.497 -> 1.571 | Improved |
| increase number of layers K from 1 to 5 (ablation) | GCN · Cora/Citeseer/Pubmed | generalization gap | Hurt |
| double hidden width from d=32 to 2d (ablation) | 5-layer GCN with normalized filter · Cora/Citeseer/Pubmed | generalization gap | Hurt |
Closest papers by meaning
- On Provable Benefits of Depth in Training Graph Convolutional Networks (NeurIPS 2021 · similarity 0.94)
- On the Stability of Graph Convolutional Neural Networks: A Probabilistic Perspective (NeurIPS 2025 · similarity 0.93)
- Convergence and Stability of Graph Convolutional Networks on Large Random Graphs (NeurIPS 2020 · similarity 0.92)
- Simple and Deep Graph Convolutional Networks (ICML 2020 · similarity 0.92)
- Simplifying Graph Convolutional Networks (ICML 2019 · similarity 0.92)
- A Manifold Perspective on the Statistical Generalization of Graph Neural Networks (ICML 2025 · similarity 0.91)
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.
