Research map / Learning theory and algorithms / Clustering
Approximate K-Means++ in Sublinear Time
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 |
|---|---|---|---|
| K-MC2 MCMC seeding (m=20) instead of exact k-means++ seeding | k-means++ · BIGX | relative error vs k-means++: 0.05% | No clear effect |
| K-MC2 instead of HEURISTIC subsampling at same distance evaluations | seeding · BIGX | relative error vs k-means++: 0.38% -> 0.05% | Helped |
| increase MCMC chain length m from 20 to 200 (ablation) | K-MC2 · CSN | relative error vs k-means++: 63.58% -> 6.53% | Helped |
| K-MC2 vs uniform random seeding | seeding · WEB | relative error vs k-means++: 105.34% -> 0.77% | Helped |
| run k-means|| with low budget (r*l < k) | k-means|| · USGS | relative error vs k-means++: 2356.06% | Hurt |
| run k-means|| with large oversampling l=2k | k-means|| · KDD | relative error vs k-means++: -6.22% | Helped |
Closest papers by meaning
- Fast and Accurate $k$-means++ via Rejection Sampling (NeurIPS 2020 · similarity 0.93)
- Fast and Provably Good Seedings for k-Means (NeurIPS 2016 · similarity 0.92)
- k-variates++: more pluses in the k-means++ (ICML 2016 · similarity 0.90)
- Distributed and Provably Good Seedings for k-Means in Constant Rounds (ICML 2017 · similarity 0.89)
- Near-Linear Time Approximation Algorithms for k-means with Outliers (ICML 2024 · similarity 0.89)
- Analyzing $D^\alpha$ seeding for $k$-means (ICML 2024 · similarity 0.88)
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.