Research map / Learning theory and algorithms / Matrix and tensor methods
Understanding Sparse JL for Feature Hashing
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 |
|---|---|---|---|
| Increase sparse JL sparsity from 1 to 4 (ablation) | Sparse JL · News20 and Enron | failure probability: factor of four decrease | Helped |
| Increase sparse JL sparsity from 1 to 8 or 16 (ablation) | Sparse JL · News20 and Enron | failure probability: factor of ten decrease | Helped |
| Increase sparsity from 1 to 2, 4, 8, or 16 (ablation) | Block sparse JL · News20 and Enron | failure probability | Helped |
| Increase sparsity from 1 to values greater than 1 (ablation) | Block sparse JL · Synthetic binary vectors | estimated v(m, ε, δ, s) | Helped |
| Increase sparsity (ablation) | Block sparse JL · Synthetic binary vectors | estimated v(m, ε, δ, s) | Mixed |
| Increase sparsity from 8 to 16 (ablation) | Block sparse JL · News20 and Enron | failure probability | Mixed |
Closest papers by meaning
- Fully Understanding The Hashing Trick (NeurIPS 2018 · similarity 0.86)
- Johnson-Lindenstrauss Lemma Beyond Euclidean Geometry (NeurIPS 2025 · similarity 0.83)
- Sparse Dimensionality Reduction Revisited (ICML 2024 · similarity 0.82)
- Optimization Can Learn Johnson Lindenstrauss Embeddings (NeurIPS 2024 · similarity 0.80)
- Euclidean distance compression via deep random features (NeurIPS 2024 · similarity 0.80)
- Beyond Worst-Case Dimensionality Reduction for Sparse Vectors (ICLR 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.