Research map / Optimization / Implicit bias and generalization
A Convergence Analysis of Gradient Descent on Graph Neural 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 |
|---|---|---|---|
| Increase graph degree d toward n (d in 5,10,30,80,99) | Gradient descent on one-round ReLU GNN · random d-regular graphs n=100 | l2 loss convergence rate | Helped |
| Use deep linear GNN instead of ReLU GNN | Gradient descent on GNN · random d-regular graphs n=100 | l2 loss convergence rate | Helped |
| Vary depth L in 1,2,3 | deep linear GNN · random d-regular graphs n=100 | l2 loss convergence rate | No clear effect |
Closest papers by meaning
- Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth (ICML 2021 · similarity 0.91)
- Learning Graph Neural Networks with Approximate Gradient Descent (AAAI 2021 · similarity 0.91)
- How Graph Neural Networks Learn: Lessons from Training Dynamics (ICML 2024 · similarity 0.91)
- On Vanishing Gradients, Over-Smoothing, and Over-Squashing in GNNs: Bridging Recurrent and Graph Learning (NeurIPS 2025 · similarity 0.90)
- On the Interplay between Graph Structure and Learning Algorithms in Graph Neural Networks (ICML 2025 · similarity 0.90)
- Almost Surely Asymptotically Constant Graph Neural Networks (NeurIPS 2024 · similarity 0.89)
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.