Research map / Learning theory and algorithms / Optimal transport
Optimal Transport: Fast Probabilistic Approximation with Exact Solvers
Max Sommerfeld, Jörn Schrieber, Yoav Zemel, Axel Munk
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 |
|---|---|---|---|
| increase the number of sampled points S (ablation) | subsampling Wasserstein approximation · Images and point clouds | relative error and runtime: S = 4000: about 3% error with two orders of magnitude speedup; S = 500: under 10% error with over four orders of magnitu | Improved |
| increase number of repetitions B (ablation) | subsampling Wasserstein approximation · Images and point clouds | relative error | No clear effect |
| use Cauchy Density images instead of White Noise images | subsampling Wasserstein approximation · Image classes | relative error | Improved |
| approximate Sinkhorn distance instead of exact Wasserstein | subsampling approximation · Images | relative error | Improved |
| switch back-end solver among exact Wasserstein solvers (ablation) | subsampling approximation · Images | relative error | No clear effect |
| increase image resolution | subsampling approximation · Images | relative runtime and error | Mixed |
Closest papers by meaning
- A Graph Theoretic Additive Approximation of Optimal Transport (NeurIPS 2019 · similarity 0.90)
- Computational Optimal Transport: Complexity by Accelerated Gradient Descent Is Better Than by Sinkhorn’s Algorithm (ICML 2018 · similarity 0.89)
- A Combinatorial Algorithm for the Semi-Discrete Optimal Transport Problem (NeurIPS 2024 · similarity 0.89)
- Geometry-Aware Optimal Transport: Fast Intrinsic Dimension and Wasserstein Distance Estimation (NeurIPS 2026 · similarity 0.88)
- Computing Optimal Transport Maps and Wasserstein Barycenters Using Conditional Normalizing Flows (ICML 2025 · similarity 0.88)
- Learning Ultrametric Trees for Optimal Transport Regression (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.
