Research map / Learning theory and algorithms / Sample complexity
Adaptive Newton Method for Empirical Risk Minimization to Statistical Accuracy
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 |
|---|---|---|---|
| Use Ada Newton with sample size doubled at each iteration instead of Newton with backtracking | ERM on Protein homology · Protein homology | passes to reach statistical accuracy 1/N: 7.5 passes -> 2.4 passes | Helped |
| Use Ada Newton instead of SAGA | ERM on Protein homology · Protein homology | passes to reach statistical accuracy 1/N: 10 passes -> 2.4 passes | Helped |
| Use Ada Newton instead of SGD | ERM on Protein homology · Protein homology | passes to reach statistical accuracy 1/N: 25 passes -> 2.4 passes | Helped |
| Use Ada Newton instead of SAGA | ERM on Protein homology · Protein homology | runtime to reach statistical accuracy 1/N: 62 seconds; Ada Newton in less than 25 seconds | Helped |
| Use Ada Newton instead of Newton with backtracking | ERM on Protein homology · Protein homology | runtime to reach statistical accuracy 1/N | Helped |
Closest papers by meaning
- Exploiting Local Convergence of Quasi-Newton Methods Globally: Adaptive Sample Size Approach (NeurIPS 2021 · similarity 0.92)
- First-Order Adaptive Sample Size Methods to Reduce Complexity of Empirical Risk Minimization (NeurIPS 2017 · similarity 0.91)
- Risk Minimization from Adaptively Collected Data: Guarantees for Supervised and Policy Learning (NeurIPS 2021 · similarity 0.87)
- On the Asymptotic Distribution of the Minimum Empirical Risk (ICML 2024 · similarity 0.86)
- Asymptotic Theory of Iterated Empirical Risk Minimization, with Applications to Active Learning (ICML 2026 · similarity 0.86)
- On the Efficiency of ERM in Feature Learning (NeurIPS 2024 · 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.