Research map / Optimization / Nonconvex and smooth optimization
High Probability Complexity Bounds for Line Search Based on Stochastic Oracles
Billy Jin, Katya Scheinberg, Miaolan Xie
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 stochastic line search with estimated noise tolerance eps_f (ALOE) instead of SLS (eps_f=0, same mini-batch while backtracking) | ALOE · PMLB 64 datasets | average best training loss: 62 out of 64 | Improved |
| use ALOE instead of full gradient line search | ALOE · PMLB 64 datasets | average best training loss | Improved |
| ALOE vs SLS and full gradient line search on test accuracy | ALOE · PMLB 80:20 splits | test accuracy | No clear effect |
| vary the choice of eps_f (ablation) | ALOE · PMLB | training loss | No clear effect |
Closest papers by meaning
- Probabilistic Line Searches for Stochastic Optimization (JMLR 2017 · similarity 0.91)
- Smooth Convex Optimization Using Sub-Zeroth-Order Oracles (AAAI 2021 · similarity 0.90)
- Gradient-free optimization of highly smooth functions: improved analysis and a new algorithm (JMLR 2024 · similarity 0.89)
- Stochastic Optimization with Random Search (NeurIPS 2026 · similarity 0.88)
- Stochastic Zeroth-Order Optimization Under Heavy-Tailed Noise (NeurIPS 2026 · similarity 0.87)
- Sample complexity of stochastic optimization with integer variables (NeurIPS 2026 · similarity 0.87)
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.
