Research map / Learning theory and algorithms / Clustering
On efficient multilevel Clustering via Wasserstein distances
Viet Huynh, Nhat Ho, Nhan Dam, XuanLong Nguyen, Mikhail Yurochkin, Hung Bui, Dinh Phung
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 |
|---|---|---|---|
| add shared atoms across local measures (MWMS vs MWM) (ablation) | MWM · LabelMe clustering | NMI: 0.374 -> 0.416 | Improved |
| add shared atoms across local measures (MWMS vs MWM) (ablation) | MWM · LabelMe clustering | ARI: 0.302 -> 0.355 | Improved |
| add context (tags) observations (ablation) | MWM · LabelMe clustering | NMI: 0.374 -> 0.662 | Improved |
| add context (tags) observations (ablation) | MWMS · LabelMe clustering | NMI: 0.416 -> 0.675 | Improved |
| add context (tags) observations (ablation) | MWMS · LabelMe clustering | ARI: 0.355 -> 0.603 | Improved |
| replace MWM with MWGM (geometric median, first-order Wasserstein) | MWM · LabelMe clustering (0% noise) | NMI: 0.473 -> 0.536 | Improved |
| replace MWM with MWGM (geometric median, first-order Wasserstein) | MWM · LabelMe with 5% Gaussian noise | NMI: 0.470 -> 0.542 | Improved |
| replace MWM with MWGM (geometric median, first-order Wasserstein) | MWM · LabelMe with 0.5% Gaussian noise | ARI: 0.340 -> 0.440 | Improved |
| use MWM instead of TSK-means | TSK-means · LabelMe clustering | NMI: 0.203 -> 0.374 | Improved |
| use MWM instead of K-means | K-means · LabelMe clustering | NMI: 0.370 -> 0.374 | No clear effect |
| add 0.5% to 1% Gaussian noise to data | MWGM · LabelMe with Gaussian noise | NMI: 0.536 -> 0.493 -> 0.553 | Mixed |
Closest papers by meaning
- Multilevel Clustering via Wasserstein Means (ICML 2017 · similarity 0.99)
- Variational Wasserstein Clustering (ECCV 2018 · similarity 0.88)
- Expected Probabilistic Hierarchies (NeurIPS 2024 · 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)
- Ultrametric Cluster Hierarchies: I Want ‘em All! (NeurIPS 2025 · similarity 0.86)
- Hierarchical Overlapping Clustering on Graphs: Cost Function, Algorithm and Scalability (ICML 2025 · similarity 0.86)
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.
