Research map / Learning theory and algorithms / Clustering
Clustering Stable Instances of Euclidean k-means.
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 Algorithm rather than k-means++ for initialization | k-means · Wine, Iris, Banknote Authentication, and Letter Recognition; normalized and unnormalized | initial k-means cost: Wine: 2.426e+06 -> 2.376e+06; Wine (normalized): 65.50 -> 48.99; Iris: 86.45 -> 81.04; Iris (normalized): 7.676 -> 7.035 | Helped |
| Initialize Lloyd's with Algorithm rather than k-means++ | Lloyd's algorithm · Wine, Iris, Banknote Authentication, and Letter Recognition; normalized and unnormalized | k-means cost after Lloyd's: Wine: 2.371e+06 vs 2.371e+06; Wine (normalized): 48.99 vs 48.95; Iris: 78.95 vs 78.94; Iris (normalized): 6.998 vs 6.998 | Mixed |
Closest papers by meaning
- Average Sensitivity of Euclidean k-Clustering (NeurIPS 2022 · similarity 0.90)
- Massively Parallel $k$-Means Clustering for Perturbation Resilient Instances (ICML 2022 · similarity 0.88)
- Online Clustering with Nearly Optimal Consistency (ICLR 2025 · similarity 0.87)
- Relax and Merge: A Simple Yet Effective Framework for Solving Fair $k$-Means and $k$-sparse Wasserstein Barycenter Problems (ICLR 2025 · similarity 0.87)
- Constant Approximation for Individual Preference Stable Clustering (NeurIPS 2023 · similarity 0.87)
- The Catastrophic Failure of *the* k-Means Algorithm in High Dimensions, and How Hartigan's Algorithm Avoids It (ICML 2026 · similarity 0.87)
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.