Research map / Learning theory and algorithms / Matrix and tensor methods
Thunder: a Fast Coordinate Selection Solver for Sparse Learning
Shaogang Ren, Weijie Zhao, Ping Li
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 |
|---|---|---|---|
| use Thunder instead of Celer and Blitz on simulated LASSO | Thunder · Simulated data n=10^4, p=10^8 | running time | Improved |
| use smaller lambda (0.3 lambda_max vs 0.7) when comparing Thunder to existing safe solvers | Thunder · Simulated data | running time | Improved |
| use Thunder instead of Celer and Blitz on E2006-log1p | Thunder · Finance data E2006-log1p, gap 1.0E-7 and 1.0E-9 | running time | Improved |
| tighten solution accuracy from 1.0E-4 to 1.0E-11 at lambda 0.005 lambda_max | Celer · Finance data | running time | Hurt |
| tighten solution accuracy from 1.0E-4 to 1.0E-11 at lambda 0.005 lambda_max | Thunder · Finance data | running time | No clear effect |
| use Thunder's active set instead of Celer's growing working set | Thunder · Finance data, lambda 0.005 lambda_max | active/working set size | Improved |
| use Thunder instead of Celer on a 10-value LASSO path with warm start | Thunder · KDD2010 and URL, gaps 1.0E-5 and 1.0E-8 | running time | Improved |
| raise path solution accuracy from 1.0E-5 to 1.0E-8 | Thunder vs Celer · KDD2010 and URL LASSO path | relative running time | Improved |
Closest papers by meaning
- Iterative Regularization with k-support Norm: An Important Complement to Sparse Recovery (AAAI 2024 · similarity 0.85)
- Beyond L1: Faster and Better Sparse Models with skglm (NeurIPS 2022 · similarity 0.85)
- Adaptive Forward Stepwise: A Method for High Sparsity Regression (JMLR 2026 · similarity 0.85)
- Structured Sparsity Optimization with Non-Convex Surrogates of $\ell _{2,0}$-Norm: A Unified Algorithmic Framework (TPAMI 2022 · similarity 0.85)
- Iterative Hard Thresholding with Adaptive Regularization: Sparser Solutions Without Sacrificing Runtime (ICML 2022 · similarity 0.85)
- L0Learn: A Scalable Package for Sparse Learning using L0 Regularization (JMLR 2023 · 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.
