Research map / Learning theory and algorithms / Clustering
Fast k-means with accurate bounds
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 |
|---|---|---|---|
| Simplify elk to selk (remove final filter) (ablation) | elk · 18 experiments, d>=20 | runtime | Helped |
| Simplify yin to syin (ablation) | yin · 44 experiments | runtime | Helped |
| Replace Annular with Exponion | ann · 22 low-dim experiments | runtime | Helped |
| Use ns-bounds instead of sn-bounds | fastest sn-algorithm · 44 {dataset,k} experiments | runtime | Helped |
| Use ns-bounds instead of sn-bounds | fastest sn-algorithm · 44 {dataset,k} experiments | total distance calculations | Mixed |
| Deactivate BLAS (ablation) | syin · low-dim datasets | speedup of syin over yin: 10% | Hurt |
| Run with four threads instead of one | own-syin-ns · datasets xii-xxii, k=100 | runtime ratio: 0.27 | Helped |
| Own optimised implementations vs published ones | elk, ham, ann, yin · 22 datasets | runtime ratio | Helped |
Closest papers by meaning
- Angle K-Means (ICLR 2026 · similarity 0.90)
- Speeding up k-means by approximating Euclidean distances via block vectors (ICML 2016 · similarity 0.87)
- Fast and Accurate $k$-means++ via Rejection Sampling (NeurIPS 2020 · similarity 0.87)
- Statistical and Computational Trade-Offs in Kernel K-Means (NeurIPS 2018 · similarity 0.86)
- LSDS++ : Dual Sampling for Accelerated k-means++ (ICML 2023 · similarity 0.86)
- Near-Linear Time Approximation Algorithms for k-means with Outliers (ICML 2024 · similarity 0.86)
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.