Research map / Learning theory and algorithms / Optimal transport
The Unbalanced Gromov Wasserstein Distance: Conic Formulation and Relaxation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 UGW (soft KL marginal penalties) instead of partial GW | PU learning on Caltech office · Cross-feature and cross-domain PU tasks | accuracy (%): surf-C -> surf-A (0.2): 66.0 -> 76.4; surf-C -> decaf-A (0.2): 87.9 -> 82.4 | Mixed |
| Apply UGW on top of FLB initialization | PU learning across different feature spaces · surf-C -> decaf-W, prior 0.2 | accuracy (%): 81.5 -> 89.9 | Helped |
| Apply UGW/PGW on top of PW initialization when source and target are same domain | PU learning · decaf-C -> decaf-C, prior 0.1 | accuracy (%): 93.9 -> 86.8 | Hurt |
Closest papers by meaning
- Partial Gromov-Wasserstein Metric (ICLR 2025 · similarity 0.91)
- Linear Partial Gromov-Wasserstein Embedding (ICLR 2025 · similarity 0.90)
- Joint Metric Space Embedding by Unbalanced Optimal Transport with Gromov–Wasserstein Marginal Penalization (ICML 2025 · similarity 0.89)
- Towards Understanding Gradient Dynamics of the Sliced-Wasserstein Distance via Critical Point Analysis (ICML 2025 · similarity 0.88)
- Semidefinite Relaxations of the Gromov-Wasserstein Distance (NeurIPS 2024 · similarity 0.88)
- Statistical and Computational Guarantees of Kernel Max-Sliced Wasserstein Distances (ICML 2025 · similarity 0.88)
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.