Research map / Label-efficient and robust learning / Domain adaptation and generalization
Variational Wasserstein Clustering
Liang Mi, Wen Zhang, Xianfeng Gu, Yalin Wang
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 |
|---|---|---|---|
| adapt source samples to the target with VWC before training the SVM | RBF SVM trained on source · Synthetic 2D domain adaptation, RBF kernel | accuracy: 59.80 -> 99.31 | Improved |
| replace k-means++ without recentering by VWC | k-means++ · Synthetic 2D domain adaptation, RBF kernel | accuracy: 50.00 -> 99.31 | Improved |
| recenter source and target domains before k-means++ (ablation) | k-means++ · Synthetic 2D domain adaptation, RBF kernel | accuracy: 50.00 -> 99.12 | Improved |
| use VWC instead of recentered k-means++ (linear kernel) | k-means++ (recentered) · Synthetic 2D domain adaptation, linear kernel | accuracy: 97.88 -> 98.56 | Improved |
| use VWC instead of D2 (linear kernel) | D2 · Synthetic 2D domain adaptation, linear kernel | accuracy: 95.85 -> 98.56 | Improved |
| use VWC instead of D2 (RBF kernel) | D2 · Synthetic 2D domain adaptation, RBF kernel | accuracy: 99.25 -> 99.31 | No clear effect |
| use VWC instead of JDOT (linear kernel) | JDOT · Synthetic 2D domain adaptation, linear kernel | accuracy: 99.03 -> 98.56 | Hurt |
| use VWC instead of JDOT (RBF kernel) | JDOT · Synthetic 2D domain adaptation, RBF kernel | accuracy: 99.23 -> 99.31 | No clear effect |
| use VWC representation instead of PCA for 3D brain MRI features | PCA + linear SVM · AD vs NC classification on 100 MRI images (5-fold CV) | accuracy | Improved |
Closest papers by meaning
- Transport Clustering: Solving Low-Rank Optimal Transport via Clustering (ICML 2026 · similarity 0.89)
- Multilevel Clustering via Wasserstein Means (ICML 2017 · similarity 0.89)
- On efficient multilevel Clustering via Wasserstein distances (JMLR 2021 · similarity 0.88)
- THESAURUS: Contrastive Graph Clustering by Swapping Fused Gromov-Wasserstein Couplings (AAAI 2025 · 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 0.86)
- Competitively Consistent Clustering (ICML 2025 · similarity 0.85)
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.
