Research map / Learning theory and algorithms / Clustering
Kernel-estimated Nonparametric Overlap-Based Syncytial Clustering
Israel A. Almodóvar-Rivera, Ranjan Maitra
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 KNOB-SynC instead of K-mH (second best) | syncytial clustering · Aggregation (2D, n=788) | adjusted Rand index: 0.95 -> 0.98 | Improved |
| use KNOB-SynC instead of DEMP/DEMP+ | syncytial clustering · Aggregation (2D, n=788) | adjusted Rand index: 0.91 -> 0.98 | Improved |
| use KNOB-SynC instead of MMC | syncytial clustering · Aggregation (2D, n=788) | adjusted Rand index: 0.8 -> 0.98 | Improved |
| provide kernel k-means with the true number of groups | kernel k-means · Aggregation (2D, n=788) | adjusted Rand index: 0.24 | Hurt |
| treat DBSCAN* outliers as singleton clusters | DBSCAN* · Aggregation (2D, n=788) | adjusted Rand index: 0.58 | Hurt |
| KNOB-SynC on Path-based, terminating early | KNOB-SynC · Path-based (2D, n=300) | adjusted Rand index: 0.55 | Hurt |
| use KNOB-SynC instead of spectral clustering on Path-based | syncytial clustering · Path-based (2D, n=300) | adjusted Rand index: 0.55 -> 0.72 | Hurt |
| syncytial merging versus plain k-means-like methods on regular clusters | syncytial methods · Simplex-7 Gaussian clusters | adjusted Rand index: 0.97 | No clear effect |
| DBSCAN* on regular Gaussian clusters | DBSCAN* · Simplex-7 Gaussian clusters | adjusted Rand index: 0.03 | Hurt |
| use EAC instead of KNOB-SynC | KNOB-SynC · E. coli protein localization (n=324, 5 classes) | adjusted Rand index: 0.72 -> 0.77 | Improved |
| run methods on first kernel PCs instead of raw data (dimension >10) (ablation) | all methods except GSL-NN · higher-dimensional datasets | clustering performance | Improved |
Closest papers by meaning
- Non-Exhaustive, Overlapping Clustering (TPAMI 2018 · similarity 0.87)
- Near-Optimal $k$-Clustering in the Sliding Window Model (NeurIPS 2023 · similarity 0.85)
- Global-Local Dirichlet Processes for Clustering Grouped Data in the Presence of Group-Specific Idiosyncratic Variables (ICML 2025 · similarity 0.85)
- Hierarchical Overlapping Clustering on Graphs: Cost Function, Algorithm and Scalability (ICML 2025 · similarity 0.85)
- Simple, Scalable and Effective Clustering via One-Dimensional Projections (NeurIPS 2023 · similarity 0.84)
- Achieving Optimal Clustering in Gaussian Mixture Models with Anisotropic Covariance Structures (NeurIPS 2024 · similarity 0.83)
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.
