Research map / Graph learning / GNN architectures
How Wide and How Deep? Mitigating Over-squashing of GNNs via Channel Capacity Constrained Estimation
Zinuo You, Jin Zheng, John Cartlidge
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 |
|---|---|---|---|
| choose hidden dimension and depth with C3E instead of heuristic defaults | GCN · Cora | accuracy: 0.808 -> 0.837 | Improved |
| choose hidden dimension and depth with C3E | JacobiConv · Chameleon | accuracy: 0.423 -> 0.469 | Improved |
| choose hidden dimension and depth with C3E | APPNP · ogbn-papers100M | accuracy: 0.637 -> 0.681 | Improved |
| choose dimensions with C3E | GPRGNN · Chameleon | accuracy: 0.348 -> 0.379 | Improved |
| increase propagation depth from 1 to 7 with fixed width 16 (ablation) | GCN · Citeseer | accuracy: 0.707 -> 0.550 | Hurt |
| increase depth from 1 to 7 with C3E-estimated widths (ablation) | GCN · Citeseer | accuracy: 0.716 -> 0.714 (L=2 to 7) | No clear effect |
| increase hidden dimension (ablation) | spectral GNNs · theory / Table 3 | representation compression ratio | Improved |
Closest papers by meaning
- PANDA: Expanded Width-Aware Message Passing Beyond Rewiring (ICML 2024 · similarity 0.87)
- Mitigating Over-squashing without Rewiring: A Sheaf Effective Resistance Perspective (NeurIPS 2026 · similarity 0.87)
- Graph Neural Networks Do Not Always Oversmooth (NeurIPS 2024 · similarity 0.87)
- Discrete Curvature Graph Information Bottleneck (AAAI 2025 · similarity 0.86)
- On Vanishing Gradients, Over-Smoothing, and Over-Squashing in GNNs: Bridging Recurrent and Graph Learning (NeurIPS 2025 · similarity 0.86)
- The Effectiveness of Curvature-Based Rewiring and the Role of Hyperparameters in GNNs Revisited (ICLR 2025 · similarity 0.86)
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.
