AI Paper Atlas

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

JMLR 2021 · Learning theory and algorithms › Clustering · approach: Clustering approximation algorithms · 4 citations (OpenAlex)

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 changedWhereMeasuredEffect
add shared atoms across local measures (MWMS vs MWM) (ablation)MWM · LabelMe clusteringNMI: 0.374 -> 0.416Improved
add shared atoms across local measures (MWMS vs MWM) (ablation)MWM · LabelMe clusteringARI: 0.302 -> 0.355Improved
add context (tags) observations (ablation)MWM · LabelMe clusteringNMI: 0.374 -> 0.662Improved
add context (tags) observations (ablation)MWMS · LabelMe clusteringNMI: 0.416 -> 0.675Improved
add context (tags) observations (ablation)MWMS · LabelMe clusteringARI: 0.355 -> 0.603Improved
replace MWM with MWGM (geometric median, first-order Wasserstein)MWM · LabelMe clustering (0% noise)NMI: 0.473 -> 0.536Improved
replace MWM with MWGM (geometric median, first-order Wasserstein)MWM · LabelMe with 5% Gaussian noiseNMI: 0.470 -> 0.542Improved
replace MWM with MWGM (geometric median, first-order Wasserstein)MWM · LabelMe with 0.5% Gaussian noiseARI: 0.340 -> 0.440Improved
use MWM instead of TSK-meansTSK-means · LabelMe clusteringNMI: 0.203 -> 0.374Improved
use MWM instead of K-meansK-means · LabelMe clusteringNMI: 0.370 -> 0.374No clear effect
add 0.5% to 1% Gaussian noise to dataMWGM · LabelMe with Gaussian noiseNMI: 0.536 -> 0.493 -> 0.553Mixed

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy