Research map / Learning theory and algorithms / Optimal transport
Statistical bounds for entropic optimal transport: sample complexity and the central limit theorem
Gonzalo Mena, Jonathan Niles-Weed
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 the new empirical-process analysis for entropic OT on subgaussian measures, including unbounded measures | Empirical entropic OT · Subgaussian measures in R^d | expected absolute estimation error: K_d (1 + σ^{⌈5d/2⌉ + 6})/√n; Genevay et al.: K_{D,d} (1 + 1/ε^{⌊d/2⌋}) e^{D²/ε}/√n | Improved |
| Draw Q_n by adding Gaussian noise to the same P_n samples instead of using independent samples | Entropic OT entropy estimator · Gaussian-mixture entropy estimation simulation | convergence speed | Improved |
| Increase dimension or decrease the regularization parameter | Empirical entropic OT · Gaussian-mixture simulation | estimation quality | Hurt |
Closest papers by meaning
- Lower Complexity Adaptation for Empirical Entropic Optimal Transport (JMLR 2024 · similarity 0.92)
- Generalization of noisy SGD in unbounded non-convex settings (ICML 2025 · similarity 0.85)
- Generalization Bounds for Heavy-Tailed SDEs through the Fractional Fokker-Planck Equation (ICML 2024 · similarity 0.85)
- Minimax Private Estimation of Smooth Optimal-Transport Maps (NeurIPS 2026 · similarity 0.84)
- Sharp Optimality of Simple, Plug-in Estimation of the Fisher Information of a Smoothed Density (ICML 2025 · similarity 0.84)
- NETS: A Non-equilibrium Transport Sampler (ICML 2025 · 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.
