Research map / Learning theory and algorithms / Clustering
Randomly Projected Convex Clustering Model: Motivation, Realization, and Cluster Recovery Guarantees
Ziwen Wang, Yancheng Yuan, Jiaming Ma, Tieyong Zeng, Defeng Sun
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 |
|---|---|---|---|
| reduce embedding dimension from 1555 to 69 in randomly projected K-means (ablation) | RP K-means · Synthetic Gaussian mixture A (K=20, d=2000) | adjusted rand index: 0.8426 -> 0.0759 | Hurt |
| reduce embedding dimension from 1555 to 69 in randomly projected convex clustering (ablation) | RPCCM · Synthetic Gaussian mixture A (K=20, d=2000) | adjusted rand index: 1.0000 -> 1.0000 | No clear effect |
| replace K-means with convex clustering at m=69 | RP K-means vs RPCCM · Synthetic Gaussian mixture A (K=20, d=2000) | adjusted rand index: 0.0759 -> 1.0000 | Improved |
| random projection of convex clustering from d=2000 to m=1555 (ablation) | CCM · Synthetic Gaussian mixture A (K=20, d=2000) | adjusted rand index: 1.0000 -> 1.0000 | No clear effect |
| lower embedding dimension to m=8 (eps=0.95) with m=ceil(eps^-2 log n) (ablation) | RPCCM · Synthetic Gaussian mixture A, 10 random projections | perfect recovery runs: 10/10 at m=173; 10/10 at m=8 | No clear effect |
| reduce embedding dimension from 12625 to m=10 (ablation) | RPCCM · Lung cancer (56 subjects, K=4) | accuracy: 55/56 -> 55/56 | No clear effect |
| reduce embedding dimension from 12625 to m=10 (ablation) | RP 4-means · Lung cancer (56 subjects, K=4) | adjusted rand index: 0.9586 -> 0.6795 | Hurt |
| reduce embedding dimension from 12625 to m=10 (ablation) | RP 4-means · Lung cancer (56 subjects, K=4) | accuracy: 55/56 -> 43/56 | Hurt |
| use convex clustering instead of K-means at m=10 | RPCCM vs RP 4-means · Lung cancer (56 subjects, K=4) | rand index: 0.8753 -> 0.9838 | Improved |
Closest papers by meaning
- Convex Clustering: Model, Theoretical Guarantee and Efficient Algorithm (JMLR 2021 · similarity 0.90)
- Simple, Scalable and Effective Clustering via One-Dimensional Projections (NeurIPS 2023 · similarity 0.88)
- Convex Clustering Redefined: Robust Learning with the Median of Means Estimator (AAAI 2026 · similarity 0.88)
- Robust Discriminative Clustering with Sparse Regularizers (JMLR 2017 · similarity 0.87)
- Adaptive Clustering through Semidefinite Programming (NeurIPS 2017 · similarity 0.87)
- How to tell when a clustering is (approximately) correct using convex relaxations (NeurIPS 2018 · 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.
