Research map / Label-efficient and robust learning / Recognition and long-tail
Simultaneous Safe Screening of Features and Samples in Doubly Sparse Modeling
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 |
|---|---|---|---|
| alternately iterating safe feature screening and safe sample screening (simultaneous screening) | L1-penalized smoothed hinge SVC and smoothed epsilon-insensitive SVR · classification and regression on real-sim, rcv1-test, gisette, rcv1-train | total computation time for 100 solutions | Helped |
| using feature screening results to shrink the primal region for sample screening (and vice versa) | safe sample and feature screening · classification and regression experiments | screening rate | Helped |
| adding safe keeping (identifying guaranteed-active features/samples) alongside safe screening | feature and sample active-set prediction · classification and regression experiments | fraction of features/samples determined active or non-active | Helped |
| new safe feature screening for LP-based SVM that needs only arbitrary primal/dual feasible solutions instead of precise solution at different lambda | LP-based SVM · theoretical | applicability | No clear effect |
Closest papers by meaning
- Optimal Sparse Recovery with Decision Stumps (AAAI 2023 · similarity 0.86)
- Safe Element Screening for Submodular Function Minimization (ICML 2018 · similarity 0.85)
- Dynamic Sasvi: Strong Safe Screening for Norm-Regularized Least Squares (NeurIPS 2021 · similarity 0.85)
- GAP Safe Screening Rules for Sparse-Group Lasso (NeurIPS 2016 · similarity 0.85)
- A Two-Stage Approach for Learning a Sparse Model with Sharp Excess Risk Analysis (AAAI 2017 · similarity 0.85)
- On the Efficiency of ERM in Feature Learning (NeurIPS 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.