Research map / Learning theory and algorithms / Optimal transport
Learning to Match via Inverse Optimal Transport
Ruilin Li, Xiaojing Ye, Haomin Zhou, Hongyuan Zha
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 marginal relaxation via regularized Wasserstein distance (RIOT vs IOT) under noise | Inverse optimal transport · Synthetic data, sigma=8e-3 | KL to ground truth plan: 0.1476 -> 0.1365 | Improved |
| add marginal relaxation (RIOT vs IOT) for learning cost matrix | Inverse optimal transport · Synthetic data, sigma=0.08 | distance to true cost matrix: 8.4623 -> 4.7075 | Improved |
| set relaxation parameter too large (delta = 0.05) (ablation) | RIOT · Synthetic data, small noise | recovery KL | Hurt |
| tune relaxation parameter carefully (delta = 0.01, 0.001) (ablation) | RIOT · Synthetic data, small noise | recovery KL | Improved |
| use RIOT instead of itemKNN | Matching recommendation · Marriage (DHS) | RMSE (x1e-4): 2.4 -> 2.3 | Improved |
| use RIOT instead of factorization machine | Matching recommendation · Marriage (DHS) | MAE (x1e-4): 2.8 -> 1.5 | Improved |
Closest papers by meaning
- Sparsistency for inverse optimal transport (ICLR 2024 · similarity 0.88)
- Flow matching achieves almost minimax optimal convergence (ICLR 2025 · similarity 0.85)
- Low-Rank Optimal Transport through Factor Relaxation with Latent Coupling (NeurIPS 2024 · similarity 0.85)
- DiMa: Understanding the Hardness of Online Matching Problems via Diffusion Models (ICML 2025 · similarity 0.84)
- Feedback Schrödinger Bridge Matching (ICLR 2025 · similarity 0.84)
- Stochastic Matching via Local Sparsification (NeurIPS 2026 · 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.
