Research map / Learning theory and algorithms / Clustering
Convex Clustering: Model, Theoretical Guarantee and Efficient Algorithm
Defeng Sun, Kim-Chuan Toh, Yancheng Yuan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 Ssnal instead of AMA | convex clustering solver · Two Half-Moon (n=1000) | time (s): 28.27 -> 0.49 | Improved |
| use Ssnal instead of AMA | convex clustering solver · Two Half-Moon (n=5000) | time (s): AMA fails at 100,000 iterations -> 3.82 | Improved |
| use Ssnal instead of AMA | convex clustering solver · Unbalanced Gaussian (gamma=0.2) | time (s): 264.54 -> 1.15 | Improved |
| use Ssnal instead of AMA | convex clustering solver · MNIST (n=10,000) | time (s): 1753.8 -> 69.3 | Improved |
| use Ssnal instead of AMA | convex clustering solver · WINE (n=178) | time (s): 2.62 -> 0.19 | Improved |
| use Ssnal instead of AMA | convex clustering solver · Yale Face B (n=760) | time (s): 211.36 -> 35.13 | Improved |
| use Ssncg iterations instead of AMA iterations | solver · Unbalanced Gaussian (gamma=1.0) | iterations: 100000 -> 27 | Improved |
| choose gamma between theoretical bounds gamma_min=1.56e-3 and gamma_max=0.485 | weighted convex clustering · five 2D Gaussian clusters | perfect cluster recovery: perfect recovery between 1.56 x 10^-3 and 0.485 | Improved |
| use convex clustering instead of K-means | clustering model · Two Half-Moon | Rand Index: better Rand Index than K-means | Improved |
| vary gamma | AMA vs Ssnal · Two Half-Moon and MNIST | time (s): gamma in [0.2:0.2:10] | Mixed |
Closest papers by meaning
- An Efficient Semismooth Newton based Algorithm for Convex Clustering (ICML 2018 · similarity 0.94)
- Clustering by Sum of Norms: Stochastic Incremental Algorithm, Convergence and Cluster Recovery (ICML 2017 · similarity 0.92)
- Convex Clustering Redefined: Robust Learning with the Median of Means Estimator (AAAI 2026 · similarity 0.92)
- Randomly Projected Convex Clustering Model: Motivation, Realization, and Cluster Recovery Guarantees (JMLR 2025 · similarity 0.90)
- Improved Learning-augmented Algorithms for k-means and k-medians Clustering (ICLR 2023 · similarity 0.90)
- Graph-based Clustering Revisited: A Relaxation of Kernel k-Means Perspective (JMLR 2026 · 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.
