Research map / Learning theory and algorithms / Matrix and tensor methods
Blended Matching Pursuit
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 |
|---|---|---|---|
| Blend constrained gradient steps with full atom steps and dual-gap steps | Matching pursuit for sparse recovery · Sparse recovery | wall-clock convergence speed | Helped |
| Blend constrained gradient steps with full atom steps and dual-gap steps | Matching pursuit for sparse recovery · Sparse recovery test set | test error: 0.1917 -> 0.0037 | Helped |
| Use BMP instead of OMP | Sparse recovery · Sparse recovery test set | test error: 0.0036 -> 0.0037 | Hurt |
| Use BMP instead of CoGEnT | Sparse recovery · Sparse recovery test set | test error: 0.0043 -> 0.0037 | Hurt |
| Set η to 5 rather than higher values (ablation) | BMP · Sparse recovery | speed of convergence | Helped |
| Set η to 5 rather than lower values (ablation) | BMP · Sparse recovery | sparsity of iterates | Helped |
| Set η to 2 rather than η=5 (ablation) | BMP · Sparse recovery | function value: 100 seconds; 10^{-2} | Mixed |
| Use BMP instead of accMP | Matching pursuit on the random 100-dimensional toy problem · Random dictionary toy problem | sparsity of iterates: all available atoms -> half as much | Helped |
| Use BMP instead of accMP | Matching pursuit on the random 100-dimensional toy problem · Random dictionary toy problem | atoms needed to start converging significantly: ~75% -> instantaneously | Helped |
Closest papers by meaning
- On Matching Pursuit and Coordinate Descent (ICML 2018 · similarity 0.91)
- Greedy Algorithms for Cone Constrained Optimization with Convergence Guarantees (NeurIPS 2017 · similarity 0.89)
- Sparse Bayesian Learning via Stepwise Regression (ICML 2021 · similarity 0.87)
- Support Recovery for Orthogonal Matching Pursuit: Upper and Lower bounds (NeurIPS 2018 · similarity 0.87)
- Methods for Convex $(L_0,L_1)$-Smooth Optimization: Clipping, Acceleration, and Adaptivity (ICLR 2025 · similarity 0.84)
- Sketching for Convex and Nonconvex Regularized Least Squares with Sharp Guarantees (ICLR 2025 · similarity 0.83)
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.