Research map / Optimization / Implicit bias and generalization
On the Effective Number of Linear Regions in Shallow Univariate ReLU Networks: Convergence Guarantees and Implicit Bias
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 |
|---|---|---|---|
| Train with gradient flow until empirical loss is below 1/n | Single-hidden-layer univariate ReLU network · Teacher-labeled univariate binary classification | linear regions: at most 32 r + 67 | Helped |
| Use width at least floor(1.3r) rather than a narrower width | Gradient flow on the univariate ReLU network · Constructed target distribution and teacher network | population loss: \geq\frac14\p{1-0.75\alpha} | Mixed |
Closest papers by meaning
- Gradient Dynamics of Shallow Univariate ReLU Networks (NeurIPS 2019 · similarity 0.92)
- Stable Minima Cannot Overfit in Univariate ReLU Networks: Generalization by Large Step Sizes (NeurIPS 2024 · similarity 0.90)
- Stable Minima of ReLU Neural Networks Suffer from the Curse of Dimensionality: The Neural Shattering Phenomenon (NeurIPS 2025 · similarity 0.90)
- Implicit Bias of Mirror Flow for Shallow Neural Networks in Univariate Regression (ICLR 2025 · similarity 0.89)
- Simplicity Bias and Optimization Threshold in Two-Layer ReLU Networks (ICML 2025 · similarity 0.89)
- Alternating Gradient Flows: A Theory of Feature Learning in Two-layer Neural Networks (NeurIPS 2025 · similarity 0.88)
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.