Research map / Optimization / Nonconvex and smooth optimization
Adaptive SGD with Polyak stepsize and Line-search: Robust Convergence and Variance Reduction
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 |
|---|---|---|---|
| replace SPS with AdaSPS (accumulated function-value differences in the stepsize) | SPS on synthetic non-interpolated quadratics · Synthetic quadratic, non-interpolated | convergence | Helped |
| replace DecSPS with AdaSPS | DecSPS on synthetic interpolated quadratics · Synthetic quadratic, interpolated | convergence rate | Helped |
| add variance reduction (AdaSVRPS) on top of AdaSPS | AdaSPS · Synthetic quadratic, non-interpolated | convergence | Helped |
| use SPS instead of AdaSPS | AdaSPS · Synthetic quadratic, interpolated | convergence | Mixed |
| compare AdaSPS/AdaSVRPS with Adam and SPS on the duke dataset (n=48, d=7129, interpolated) | AdaSPS · LIBSVM duke | convergence speed | Hurt |
| compare AdaSPS/AdaSVRPS with seven optimizers on larger datasets | AdaSPS, AdaSVRPS · LIBSVM rcv1, ijcnn, w8a | convergence | Helped |
| compare three VR methods (SVRG, AdaSVRG, AdaSVRPS) when interpolation is not satisfied | AdaSVRPS · LIBSVM datasets | convergence | No clear effect |
Closest papers by meaning
- New Perspectives on the Polyak Stepsize: Surrogate Functions and Negative Results (NeurIPS 2025 · similarity 0.88)
- Fast Last-Iterate Convergence of SGD in the Smooth Interpolation Regime (NeurIPS 2025 · similarity 0.88)
- Taking the Road Less Scheduled with Adaptive Polyak Steps (NeurIPS 2026 · similarity 0.88)
- Stochastic Polyak Step-sizes and Momentum: Convergence Guarantees and Practical Performance (ICLR 2025 · similarity 0.87)
- Universality of AdaGrad Stepsizes for Stochastic Optimization: Inexact Oracle, Acceleration and Variance Reduction (NeurIPS 2024 · similarity 0.87)
- Non-convex Stochastic Composite Optimization with Polyak Momentum (ICML 2024 · 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.