Research map / Learning theory and algorithms / Sample complexity
Fully Understanding The Hashing Trick
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 target dimension to m ≥ 2/(ε²δ) | Feature hashing · Empirical parameter range | ν: ν = 1 when m ≥ 2/(ε²δ) | Helped |
| Use target dimension just below 2/(ε²δ) | Feature hashing · Empirical parameter range | ν: ν < 1 while m = (2 − γ)/(ε²δ) | Hurt |
| Apply feature hashing to vectors with ||x||₀ > 7 in the tested parameter range | Feature hashing · 4 lg(1/δ)/ε² ≤ m < 2/(ε²δ) | ν-to-theoretical-bound ratio: at least 0.725 | Helped |
| Apply feature hashing to very sparse vectors (||x||₀ ≤ 7) in the tested parameter range | Feature hashing · 4 lg(1/δ)/ε² ≤ m < 2/(ε²δ) | ν-to-theoretical-bound ratio: at least 0.6; at least 0.725 except for very sparse vectors (||x||₀ ≤ 7) | Mixed |
Closest papers by meaning
- Practical Hash Functions for Similarity Estimation and Dimensionality Reduction (NeurIPS 2017 · similarity 0.87)
- Optimal Densification for Fast and Accurate Minwise Hashing (ICML 2017 · similarity 0.86)
- A Lower Bound of Hash Codes' Performance (NeurIPS 2022 · similarity 0.86)
- Understanding Sparse JL for Feature Hashing (NeurIPS 2019 · similarity 0.86)
- Tricking the Hashing Trick: A Tight Lower Bound on the Robustness of CountSketch to Adaptive Inputs (AAAI 2023 · similarity 0.85)
- Optimizing affinity-based binary hashing using auxiliary coordinates (NeurIPS 2016 · 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.