Research map / Learning theory and algorithms / Matrix and tensor methods
Random extrapolation for primal-dual coordinate descent
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 PURE-CD instead of SPDHG | PURE-CD · Lasso, rcv1 (density 0.16%) | convergence speed | Helped |
| Use PURE-CD instead of Vu-Condat-CD | PURE-CD · Sparse Lasso and ridge regression datasets | convergence speed | Helped |
| Use PURE-CD instead of SPDHG or Vu-Condat-CD | PURE-CD · Moderately sparse Lasso and ridge regression datasets | convergence speed | Helped |
| Use PURE-CD instead of Vu-Condat-CD | PURE-CD · Dense Lasso and ridge regression datasets | convergence speed | Helped |
| Use agnostic step sizes rather than step sizes incorporating the strong-convexity parameter (ablation) | PURE-CD · Strongly convex ridge regression datasets | convergence speed | Mixed |
| Incorporate the strong-convexity parameter lambda in the step sizes (ablation) | PURE-CD · Strongly convex ridge regression datasets where agnostic PURE-CD slows as lambda gets smaller | convergence speed | Helped |
Closest papers by meaning
- Smooth Primal-Dual Coordinate Descent Algorithms for Nonsmooth Convex Optimization (NeurIPS 2017 · similarity 0.91)
- Doubly Greedy Primal-Dual Coordinate Descent for Sparse Empirical Risk Minimization (ICML 2017 · similarity 0.90)
- Accelerated Cyclic Coordinate Dual Averaging with Extrapolation for Composite Convex Optimization (ICML 2023 · similarity 0.89)
- Nonasymptotic Analysis of Stochastic Gradient Descent with the Richardson–Romberg Extrapolation (ICLR 2025 · similarity 0.87)
- Provable Benefit of Random Permutations over Uniform Sampling in Stochastic Coordinate Descent (ICML 2025 · similarity 0.87)
- Tighter Convergence Bounds for Shuffled SGD via Primal-Dual Perspective (NeurIPS 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.