Research map / Optimization / Nonconvex and smooth optimization
Gradient Projection Iterative Sketch for Large-Scale Constrained Least-Squares
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 full accelerated projected gradient with GPIS (sketched gradient projection with Count-sketch and line search) for constrained LS | GPIS · Syn1/Syn2/Syn3/Magic04 constrained LS (wall-clock time) | wall-clock time to reach relative error | Helped |
| Use sketched deterministic iterations instead of mini-batch SAGA | GPIS / Acc-GPIS · Syn1, Syn2, Magic04 (wall-clock time) | wall-clock time | Helped |
| Use mini-batch SAGA (b=10) in place of sketched methods on constrained problems with costly projection | SAGA b=10 · Syn3 nuclear-norm constrained (wall-clock time) | wall-clock time | Hurt |
| Full accelerated PGD rather than sketched methods on Syn3 (nuclear-norm) | Acc-PGD · Syn3 low-rank (wall-clock time) | wall-clock time | Helped |
| Run GPCS initialization loop (classical sketch warm start) before GPIHS iterations | GPIS · regression on real data sets | convergence speed | No clear effect |
| Add Nesterov acceleration with gradient restart to GPIS (Acc-GPIS) | Acc-GPIS · theory: outer-loop rate | convergence rate | Helped |
Closest papers by meaning
- Iterative Double Sketching for Faster Least-Squares Optimization (ICML 2022 · similarity 0.90)
- Optimal Iterative Sketching Methods with the Subsampled Randomized Hadamard Transform (NeurIPS 2020 · similarity 0.89)
- Sketching for Convex and Nonconvex Regularized Least Squares with Sharp Guarantees (ICLR 2025 · similarity 0.88)
- Can Gaussian Sketching Converge Faster on a Preconditioned Landscape? (ICML 2024 · similarity 0.88)
- Distributed Least Squares in Small Space via Sketching and Bias Reduction (NeurIPS 2024 · similarity 0.87)
- Fundamental Bias in Inverting Random Sampling Matrices with Application to Sub-sampled Newton (ICML 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.