Research map / Learning theory and algorithms / Matrix and tensor methods
Simplex Constrained Sparse Optimization via Tail Screening
Peng Chen, Jin Zhu, Junxian Zhu, Xueqin Wang
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 |
|---|---|---|---|
| add tail screening (ablation) | projected gradient on simplex least squares · Synthetic (n,p,s*)=(500,500,5) | support accuracy: 0.13 -> 0.97 | Improved |
| add tail screening (ablation) | projected gradient on simplex least squares · Synthetic (n,p,s*)=(500,1000,10) | support accuracy: 0.11 -> 0.98 | Improved |
| add tail screening (ablation) | projected gradient on simplex least squares · Synthetic (n,p,s*)=(1000,1000,20) | support accuracy: 0.12 -> 0.99 | Improved |
| add tail screening (ablation) | projected gradient on simplex least squares · Synthetic (n,p,s*)=(1000,2000,30) | support accuracy: 0.08 -> 0.97 | Improved |
| add tail screening (ablation) | projected gradient on simplex least squares · Synthetic (n,p,s*)=(1000,2000,30) | runtime (s): 31.11 -> 0.57 | Improved |
| add tail screening (ablation) | projected gradient on simplex least squares · Synthetic (n,p,s*)=(500,500,5) | sparsity (l0 norm): 148.44 -> 5.16 | Improved |
| add tail screening (ablation) | projected gradient on simplex least squares · Synthetic (n,p,s*)=(500,1000,10) | l2 error: 0.11 -> 0.05 | Improved |
| use PERMITS instead of H-Lasso | PERMITS · DJIA constituents detection | accuracy (%): 34% -> 93% | Improved |
| use PERMITS instead of IHT* | PERMITS · DJIA constituents detection | l2 error: 0.226 -> 0.038 | Improved |
| use PERMITS instead of H-Lasso | PERMITS · DJIA constituents detection | l2 error: 0.040 -> 0.038 | Improved |
| vary tolerance epsilon among 1e-3, 1e-4, 1e-5 times sqrt(log p/n) (ablation) | PERMITS · Synthetic simplex regression (Appendix A) | accuracy, l2 error, runtime | No clear effect |
Closest papers by meaning
- Screening rules for Lasso with non-convex Sparse Regularizers (ICML 2019 · similarity 0.88)
- Sparse Convex Optimization via Adaptively Regularized Hard Thresholding (ICML 2020 · similarity 0.86)
- A Feasible Level Proximal Point Method for Nonconvex Sparse Constrained Optimization (NeurIPS 2020 · similarity 0.86)
- Gradient Compressed Sensing: A Query-Efficient Gradient Estimator for High-Dimensional Zeroth-Order Optimization (ICML 2024 · similarity 0.86)
- Optimization over Sparse Support-Preserving Sets: Two-Step Projection with Global Optimality Guarantees (ICML 2025 · similarity 0.85)
- Straight-Through Meets Sparse Recovery: the Support Exploration Algorithm (ICML 2024 · similarity 0.85)
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.
