Research map / Optimization / Nonconvex and smooth optimization
On Matching Pursuit and 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 |
|---|---|---|---|
| Measure smoothness and level-set radius with the atomic norm | Matching Pursuit and steepest coordinate descent · Convex smooth objectives | function suboptimality: \frac{2L_{\mathcal{A}}R_{\mathcal{A}}^2}{t+2} versus \frac{2L_2nR_2^2}{t+2} | Helped |
| Use an exact steepest direction instead of a random direction | Matching Pursuit versus Random Pursuit · Convex smooth objectives | convergence speed | Helped |
| Accelerate the pursuit updates | Matching Pursuit and Random Pursuit · Synthetic sparse signal and Urban HDI hyperspectral data | loss | Helped |
| Use steepest rather than random updates | Matching Pursuit and Random Pursuit · Urban HDI hyperspectral data | loss | Mixed |
| Use greedy updates for x and random atom updates for v | Accelerated Matching Pursuit · Convex objectives | expected function suboptimality: \mathcal{O}(1/t^2) | Helped |
| Use steepest coordinate updates instead of randomized coordinate updates | Accelerated coordinate descent · Convex objectives | convergence rate | Helped |
Closest papers by meaning
- Blended Matching Pursuit (NeurIPS 2019 · similarity 0.91)
- First-Order Algorithms Converge Faster than $O(1/k)$ on Convex Problems (ICML 2019 · similarity 0.89)
- Accelerating Greedy Coordinate Descent Methods (ICML 2018 · similarity 0.89)
- Directional Smoothness and Gradient Methods: Convergence and Adaptivity (NeurIPS 2024 · similarity 0.87)
- Fundamental Benefit of Alternating Updates in Minimax Optimization (ICML 2024 · similarity 0.86)
- Accelerated Distance-adaptive Methods for Hölder Smooth and Convex Optimization (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.