Research map / Learning theory and algorithms / Optimal transport
Low-rank Optimal Transport: Approximation, Statistics and Debiasing
Meyer Scetbon, marco cuturi
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 |
|---|---|---|---|
| use debiased DLOT instead of biased LOT as the gradient-flow objective (ablation) | low-rank OT gradient flow · Gaussian to moon-shape distribution, r=100, 1000 samples | recovery of target distribution | Improved |
| increase the rank r | plug-in LOT/DLOT estimator · mixture of 10 anisotropic Gaussians | statistical convergence rate | Hurt |
| increase ground-space dimension d | plug-in LOT/DLOT estimator · mixture of 10 anisotropic Gaussians | statistical convergence rate | No clear effect |
| use k-means or generalized k-means initialization instead of rank-2 or random (ablation) | mirror-descent LOT solver · Newsgroup20 embeddings, 50 runs | avoidance of spurious local minima / stopping criterion | Improved |
| vary initialization (k-means, generalized k-means, rank-2, random) (ablation) | mirror-descent LOT solver · Newsgroup20 embeddings | final cost value | No clear effect |
| use shortest-path graph cost instead of squared Euclidean cost for LOT clustering (ablation) | LOT-based clustering · 6 scikit-learn synthetic datasets | cluster recovery | Improved |
Closest papers by meaning
- Unbalanced Low-rank Optimal Transport Solvers (NeurIPS 2023 · similarity 0.92)
- Low-Rank Optimal Transport through Factor Relaxation with Latent Coupling (NeurIPS 2024 · similarity 0.92)
- Hierarchical Refinement: Optimal Transport to Infinity and Beyond (ICML 2025 · similarity 0.89)
- Sparsistency for inverse optimal transport (ICLR 2024 · similarity 0.89)
- The Sparse-Plus-Low-Rank Quasi-Newton Method for Entropic-Regularized Optimal Transport (ICML 2025 · similarity 0.88)
- Double-Bounded Optimal Transport for Advanced Clustering and Classification (AAAI 2024 · 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.
