Research map / Learning theory and algorithms / Kernels and regression theory
$\ell _{1}$-Norm Quantile Regression Screening Rule via the Dual Circumscribed Sphere
Pan Shang, Lingchen Kong
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 |
|---|---|---|---|
| increase p from 5000 to 15000 (tau=0.75, Sigma1) (ablation) | screening rule + ADMM · Simulation n=100 | speedup: 4.3763 -> 6.8307 | Improved |
| increase tau from 0.25 to 0.75 (p=15000, Sigma1) (ablation) | screening rule + ADMM · Simulation n=100 | speedup: 4.0368 -> 6.8307 | Improved |
| increase tau from 0.25 to 0.5 (p=5000, Sigma1) (ablation) | screening rule + ADMM · Simulation n=100 | speedup: 3.2157 -> 4.4298 | Improved |
| increase tau from 0.5 to 0.75 (p=5000, Sigma1) (ablation) | screening rule + ADMM · Simulation n=100 | speedup: 4.4298 -> 4.3763 | No clear effect |
| use correlated features Sigma2 instead of identity Sigma1 (p=15000, tau=0.75) (ablation) | screening rule + ADMM · Simulation n=100 | speedup: 6.8307 -> 6.5888 | No clear effect |
| use correlated features Sigma2 instead of identity Sigma1 (p=5000, tau=0.25) (ablation) | screening rule + ADMM · Simulation n=100 | speedup: 3.2157 -> 3.1522 | No clear effect |
| tau=0.75 versus other quantile levels (ablation) | screening rule · Simulation (rejection ratio) | rejection ratio | Improved |
| apply screening rule before solving | proximal ADMM · Simulation n=100 | solver time: 22.2410 -> 5.0821 | Improved |
| apply screening rule before solving | proximal ADMM · Simulation n=100, p=15000, Sigma1, tau=0.75 | solver time: 186.1347 -> 27.2497 | Improved |
| apply screening rule before solving | proximal ADMM · Real high-dimensional datasets | speedup: 2 to 23 | Improved |
| increase tau on real data (excluding Leukemia and Brain) (ablation) | screening rule + ADMM · Real datasets | speedup | Mixed |
Closest papers by meaning
- High-Dimensional Structured Quantile Regression (ICML 2017 · similarity 0.88)
- Distributed High-Dimensional Quantile Regression: Estimation Efficiency and Support Recovery (ICML 2024 · similarity 0.86)
- Conformalized Quantile Regression (NeurIPS 2019 · similarity 0.86)
- Nearly Linear Row Sampling Algorithm for Quantile Regression (ICML 2020 · similarity 0.85)
- Finite Smoothing Algorithm for High-Dimensional Support Vector Machines and Quantile Regression (ICML 2024 · similarity 0.84)
- Relaxed Quantile Regression: Prediction Intervals for Asymmetric Noise (ICML 2024 · 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.
