Research map / Learning theory and algorithms / Clustering
Improved Learning-augmented Algorithms for k-means and k-medians Clustering
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 |
|---|---|---|---|
| Select the minimum-cost contiguous subset of size (1-α)m_i independently in each coordinate | k-means clustering · k-means, label error α<1/2 | approximation factor: 1 + (α/(1-α) + 4α/((1-2α)(1-α))) | Helped |
| Extend the supported label-error range from α<1/7 to α<1/2 | learning-augmented k-means · k-means | supported label error range: α<1/7 -> α<1/2 | Helped |
| Use deterministic minimum-cost subset selection instead of the prior randomized approach | learning-augmented k-means · CIFAR-10, MNIST, PHY | clustering cost | Mixed |
| Use the sampling-and-clipping procedure with repeated random anchors and select the candidate with lowest cost | learning-augmented k-medians · CIFAR-10, MNIST, PHY | clustering cost | Helped |
| Use variable-size random samples and select the sample size with the best cost, rather than a fixed-size random subset | random-sampling k-medians baseline · CIFAR-10, MNIST | clustering cost | Helped |
| Set the number of k-medians repetitions to R=1 | learning-augmented k-medians · CIFAR-10, MNIST, PHY | clustering cost | No clear effect |
Closest papers by meaning
- Sample-and-Search: An Effective Algorithm for Learning-Augmented k-Median Clustering in High Dimensions (AAAI 2026 · similarity 0.94)
- Learning-Augmented Algorithms for $k$-median via Online Learning (NeurIPS 2025 · similarity 0.92)
- Average Sensitivity of Hierarchical $k$-Median Clustering (ICML 2025 · similarity 0.91)
- Learning-Augmented Hierarchical Clustering (ICML 2025 · similarity 0.89)
- Dynamic Algorithm for Explainable $k$-medians Clustering under $\ell_p$ Norm (NeurIPS 2025 · similarity 0.89)
- Efficient Constrained K-center Clustering with Background Knowledge (AAAI 2024 · similarity 0.89)
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.