Research map / Learning theory and algorithms / Optimal transport
Screening Sinkhorn Algorithm for Regularized Optimal Transport
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 Screenkhorn instead of Sinkhorn | WDA on the toy dataset · WDA toy dataset | running time gain: 2 | Helped |
| Use Screenkhorn instead of Sinkhorn | WDA on MNIST · WDA MNIST | running time gain: 4.5 | Helped |
| Use Screenkhorn instead of Sinkhorn | OTDA · MNIST to USPS | running time gain: from a factor of 4 to 12 | Helped |
| Use Screenkhorn instead of Sinkhorn | WDA · MNIST | classification accuracy | No clear effect |
| Use Screenkhorn instead of Sinkhorn | OTDA · MNIST to USPS | accuracy | No clear effect |
| Use Screenkhorn with a low point budget | Sinkhorn distance computation · Synthetic toy problem | running time gain: almost 2 times faster | Helped |
| Use Screenkhorn when the budget approaches the full support | Sinkhorn distance computation · Synthetic toy problem | running time | Hurt |
| Use Screenkhorn with small regularization | Sinkhorn distance computation · Synthetic toy problem | relative divergence difference | Hurt |
| Increase the screened point budget | Screenkhorn · Synthetic toy problem | marginal violations | Helped |
| Use Screenkhorn with high regularization | Sinkhorn distance computation · Synthetic toy problem | marginal approximation: eta > 1 | Helped |
Closest papers by meaning
- On Unbalanced Optimal Transport: An Analysis of Sinkhorn Algorithm (ICML 2020 · similarity 0.91)
- On Partial Optimal Transport: Revising the Infeasibility of Sinkhorn and Efficient Gradient Methods (AAAI 2024 · similarity 0.89)
- Accelerating Sinkhorn algorithm with sparse Newton iterations (ICLR 2024 · similarity 0.88)
- Safe and Sparse Newton Method for Entropic-Regularized Optimal Transport (NeurIPS 2024 · similarity 0.88)
- A Combinatorial Algorithm for the Semi-Discrete Optimal Transport Problem (NeurIPS 2024 · similarity 0.87)
- Learning Ultrametric Trees for Optimal Transport Regression (AAAI 2024 · 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.