Research map / Optimization / Implicit bias and generalization
Early-stopped neural networks are consistent
Ziwei Ji, Justin D. Li, Matus Telgarsky
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 |
|---|---|---|---|
| Train a shallow ReLU network with logistic-loss gradient descent and select an early iterate | Shallow ReLU networks on general binary classification distributions · General binary classification | population logistic risk | Improved |
| Train with early-stopped gradient descent rather than a univariate local interpolation rule | Noisy univariate binary classification · Noisy univariate distributions | population misclassification risk | Mixed |
| Use a clairvoyantly chosen early-stopping radius instead of no radius hint (ablation) | Shallow ReLU gradient descent on general noisy data · General noisy classification | network width required for excess risk O(ε): m = 1/ε^8; m = 1/ε^40/3 | Improved |
Closest papers by meaning
- When Does Gradient Descent with Logistic Loss Find Interpolating Two-Layer Networks? (JMLR 2021 · similarity 0.89)
- Generalization and Stability of Interpolating Neural Networks with Minimal Width (JMLR 2024 · similarity 0.88)
- Simplicity Bias and Optimization Threshold in Two-Layer ReLU Networks (ICML 2025 · similarity 0.87)
- Minimax Optimal Early-Stopped Gradient Descent for Gaussian Mixture Classification (NeurIPS 2026 · similarity 0.86)
- Benefits of Early Stopping in Gradient Descent for Overparameterized Logistic Regression (ICML 2025 · similarity 0.85)
- Feature Averaging: An Implicit Bias of Gradient Descent Leading to Non-Robustness in Neural Networks (ICLR 2025 · similarity 0.85)
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.
