Research map / Optimization / Nonconvex and smooth optimization
Weighted SGD for $\ell_p$ Regression with Randomized Preconditioning
Jiyan Yang, Yin-Lam Chow, Christopher Ré, Michael W. Mahoney
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 full preconditioner F=R^-1 (pwSGD-full) instead of no preconditioner (ablation) | pwSGD · Synthetic 2 (increasing condition number) | iterations to reach relative error 0.1 | Improved |
| use diagonal preconditioner instead of none (ablation) | pwSGD · Synthetic 2 (increasing condition number) | iterations to reach relative error 0.1 | Mixed |
| use pwSGD-full or pwSGD-diag instead of SGD, weighted-RK, SVRG, pwSGD-noco | l1 and l2 regression · Buzz (condition number 10^8) | solution error versus time | Improved |
| use pwSGD-full instead of RLA with algorithmic leveraging | l1 regression · Year and Buzz | time-accuracy tradeoff | Improved |
| use pwSGD instead of RLA for l2 regression | l2 regression · Year and Buzz | speed | No clear effect |
| use smaller constant step size eta=0.001 (ablation) | pwSGD-full · Synthetic sparse l2 regression | optimization error | Improved |
Closest papers by meaning
- Risk Bounds of Accelerated SGD for Overparameterized Linear Regression (ICLR 2024 · similarity 0.87)
- High-probability complexity bounds for stochastic non-convex minimax optimization (NeurIPS 2024 · similarity 0.87)
- Statistical Guarantees for High-Dimensional Stochastic Gradient Descent (NeurIPS 2025 · similarity 0.87)
- Investigating the Role of Weight Decay in Enhancing Nonconvex SGD (CVPR 2025 · similarity 0.87)
- Incremental Gradient Descent with Small Epoch Counts is Surprisingly Slow on Ill-Conditioned Problems (ICML 2025 · similarity 0.86)
- Fast Last-Iterate Convergence of SGD in the Smooth Interpolation Regime (NeurIPS 2025 · similarity 0.86)
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.
