Research map / Learning theory and algorithms / Matrix and tensor methods
Newton-Stein Method: An Optimization Method for GLMs via Stein's Lemma
Murat A. Erdogdu
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 |
|---|---|---|---|
| use Newton-Stein method instead of classical optimizers (Newton, BFGS, gradient methods) | GLM training · Synthetic S3 and S20 (n=500000, p=300) | convergence speed | Improved |
| use Newton-Stein method instead of classical optimizers | GLM training · Real datasets CT slices and Covertype | convergence speed | Improved |
| use a larger step size than 1 enabled by eigenvalue thresholding | Newton-Stein method · GLM training | convergence speed and stability | Mixed |
| sub-sample once only at start with poorly conditioned data or small sub-sample | Newton-Stein method · GLM training | convergence | Hurt |
Closest papers by meaning
- Scaled Least Squares Estimator for GLMs in Large-Scale Problems (NeurIPS 2016 · similarity 0.88)
- Scalable First-order Method for Certifying Optimal k-Sparse GLMs (ICML 2025 · similarity 0.87)
- Estimating High-dimensional Non-Gaussian Multiple Index Models via Stein’s Lemma (NeurIPS 2017 · similarity 0.86)
- Dual Extrapolation for Sparse GLMs (JMLR 2020 · similarity 0.86)
- A Unified Confidence Sequence for Generalized Linear Models, with Applications to Bandits (NeurIPS 2024 · similarity 0.85)
- Finite-Sample Performance of Gradient Descent in Logistic Regression with Gaussian Design (NeurIPS 2026 · similarity 0.84)
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.
