Research map / Learning theory and algorithms / Submodular and game algorithms
Greedy Column Subset Selection: New Bounds and Distributed Algorithms
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 |
|---|---|---|---|
| Use Distgreedy instead of random feature selection | news20.binary feature selection · news20.binary, n=500 | classification accuracy: 54.9 -> 80.2 | Helped |
| Use Distgreedy instead of random feature selection | news20.binary feature selection · news20.binary, n=1000 | classification accuracy: 59.2 -> 82.9 | Helped |
| Use Distgreedy instead of random feature selection | news20.binary feature selection · news20.binary, n=2500 | classification accuracy: 67.6 -> 85.5 | Helped |
| Use Distgreedy instead of 2-Phase | news20.binary feature selection · news20.binary, n=500 | classification accuracy: 81.8 -> 80.2 | Hurt |
| Use Distgreedy instead of 2-Phase | news20.binary feature selection · news20.binary, n=1000 | classification accuracy: 84.4 -> 82.9 | Hurt |
| Use Distgreedy instead of 2-Phase | news20.binary feature selection · news20.binary, n=2500 | classification accuracy: 87.9 -> 85.5 | Hurt |
| Use Distgreedy instead of 2-Phase | news20.binary feature selection · news20.binary, n=500 | speedup over 2-Phase: 1.0 -> 72.3 | Helped |
| Use Distgreedy instead of 2-Phase | news20.binary feature selection · news20.binary, n=1000 | speedup over 2-Phase: 1.0 -> 16.4 | Helped |
| Use Distgreedy instead of 2-Phase | news20.binary feature selection · news20.binary, n=2500 | speedup over 2-Phase: 1.0 -> 2.4 | Helped |
| Use the randomized coreset version of GREEDY rather than standard GREEDY | MNIST feature selection · MNIST | classification accuracy | No clear effect |
| Use column selection instead of PCA | MNIST feature selection · MNIST, 300 selected columns | reconstruction and classification performance: within 1% | Hurt |
Closest papers by meaning
- Multi-View Unsupervised Column Subset Selection via Combinatorial Search (Student Abstract) (AAAI 2025 · similarity 0.88)
- Linear Time Approximation Algorithm for Column Subset Selection with Local Search (NeurIPS 2024 · similarity 0.88)
- Improved Theoretically-Grounded Evolutionary Algorithms for Subset Selection with a Linear Cost Constraint (ICML 2025 · similarity 0.87)
- GIST: Greedy Independent Set Thresholding for Max-Min Diversification with Submodular Utility (NeurIPS 2025 · similarity 0.86)
- Multinoulli Extension: A Lossless Yet Effective Probabilistic Framework for Subset Selection over Partition Constraints (ICML 2025 · similarity 0.85)
- An Asymptotically Optimal Approximation Algorithm for Multiobjective Submodular Maximization at Scale (ICML 2025 · 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.