Research map / Optimization / Implicit bias and generalization
Parameters as interacting particles: long time convergence and asymptotic error scaling of 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 number of hidden units n in a wide single-layer ReLU network | SGD-trained ReLU network · 3-spin target, d=50 | empirical loss | Helped |
| increase number of hidden units n in sigmoid networks | SGD-trained sigmoid network · 3-spin target, d=10 and d=25 | loss and signed discrepancy | Helped |
| quench SGD noise with larger batch size or smaller step late in training | SGD · asymptotic error theory and ReLU experiments | representation error order | Helped |
Closest papers by meaning
- 4+3 Phases of Compute-Optimal Neural Scaling Laws (NeurIPS 2024 · similarity 0.87)
- Symmetries in Overparametrized Neural Networks: A Mean Field View (NeurIPS 2024 · similarity 0.87)
- Learning Multi-Index Models with Neural Networks via Mean-Field Langevin Dynamics (ICLR 2025 · similarity 0.87)
- Convergence Analysis of Newton's Method for Neural Networks in the Overparameterized Limit (NeurIPS 2026 · similarity 0.86)
- On the Surprising Effectiveness of Large Learning Rates under Standard Width Scaling (NeurIPS 2025 · similarity 0.86)
- Super Consistency of Neural Network Landscapes and Learning Rate Transfer (NeurIPS 2024 · 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.