Research map / Optimization / Nonconvex and smooth optimization
skscope: Fast Sparsity-Constrained Optimization in Python
Zezhi Wang, Junxian Zhu, Xueqin Wang, Jin Zhu, Huiyang Pen, Peng Chen, Anran Wang, Xiaoke Zhang
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 skscope solvers instead of cvxpy l1 relaxation | skscope solvers · Sparse learning tasks, unknown s | runtime speedup: 1-80x speedups on cvxpy | Improved |
| use skscope solvers instead of GUROBI | skscope solvers · Sparse learning tasks, known s | runtime speedup: 30-1000x speedups on GUROBI | Improved |
| use ScopeSolver instead of FoBaSolver | ScopeSolver · Sparse learning tasks | runtime speedup: around 1.5x-7x compared to FoBaSolver | Improved |
| use ScopeSolver instead of cvxpy | ScopeSolver · Linear regression (unknown s) | F1-score: 0.60 -> 0.93 | Improved |
| use ScopeSolver instead of cvxpy | ScopeSolver · Trend filtering (unknown s) | F1-score: 0.44 -> 0.76 | Improved |
| use ScopeSolver instead of cvxpy | ScopeSolver · Nonlinear feature selection (unknown s) | F1-score: 0.79 -> 0.78 | No clear effect |
| use ScopeSolver instead of GUROBI | ScopeSolver · Trend filtering (known s) | support-set accuracy: 1.00 -> 0.98 | No clear effect |
| use ScopeSolver instead of IHTSolver | ScopeSolver · Trend filtering (known s) | support-set accuracy: 0.08 -> 0.98 | Improved |
| use specialized abess instead of generic ScopeSolver | abess · Linear regression n=5000, p=10000 | runtime (s): 14.09 -> 2.19 | Improved |
| use abess instead of ScopeSolver | abess · Linear regression n=500, p=1000 | runtime (s): 0.23 -> 0.01 | Improved |
Closest papers by meaning
- On the Treatment of Optimization Problems With L1 Penalty Terms via Multiobjective Continuation (TPAMI 2021 · similarity 0.81)
- OKRidge: Scalable Optimal k-Sparse Ridge Regression (NeurIPS 2023 · similarity 0.81)
- Improving Linear System Solvers for Hyperparameter Optimisation in Iterative Gaussian Processes (NeurIPS 2024 · similarity 0.81)
- Smoothing Proximal Gradient Methods for Nonsmooth Sparsity Constrained Optimization: Optimality Conditions and Global Convergence (ICML 2024 · similarity 0.80)
- CONGO: Compressive Online Gradient Optimization (ICLR 2025 · similarity 0.80)
- Sharper Convergence Rates for Nonconvex Optimisation via Reduction Mappings (NeurIPS 2025 · similarity 0.79)
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.
