Research map / Learning theory and algorithms / Optimal transport
Differential Properties of Sinkhorn Approximation for Learning with Wasserstein Distance
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 the sharp Sinkhorn distance instead of the regularized Sinkhorn distance for barycenter estimation | Sinkhorn barycenter estimation · Random measures with sparse support (1%, 2%, 10%, and 50%) | Wasserstein barycenter functional improvement: 14.914 \pm 0.076 (1%); 12.482 \pm 0.135 (2%); 2.736\pm 0.569 (10%); 0.258 \pm 0.012 (50%) | Helped |
| Use sharp rather than regularized Sinkhorn loss for reconstruction | Structured prediction estimator · Google QuickDraw reconstruction | Classification error | Helped |
| Use Hellinger loss instead of regularized Sinkhorn loss | Structured prediction estimator · Google QuickDraw reconstruction, 4 classes | Classification error | Helped |
Closest papers by meaning
- Accelerating Sinkhorn algorithm with sparse Newton iterations (ICLR 2024 · similarity 0.91)
- Towards Understanding Gradient Dynamics of the Sliced-Wasserstein Distance via Critical Point Analysis (ICML 2025 · similarity 0.89)
- Safe and Sparse Newton Method for Entropic-Regularized Optimal Transport (NeurIPS 2024 · similarity 0.89)
- Mirror and Preconditioned Gradient Descent in Wasserstein Space (NeurIPS 2024 · similarity 0.88)
- Geometry-Aware Optimal Transport: Fast Intrinsic Dimension and Wasserstein Distance Estimation (NeurIPS 2026 · similarity 0.87)
- The Sparse-Plus-Low-Rank Quasi-Newton Method for Entropic-Regularized Optimal Transport (ICML 2025 · similarity 0.87)
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.