Research map / Learning theory and algorithms / Optimal transport
Entropic Neural Optimal Transport via Diffusion Processes
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 |
|---|---|---|---|
| increase entropy parameter eps from 0 to 10 (ablation) | ENOT · Colored MNIST | LPIPS variability: 0 -> 5.3\cdot10^{-3} -> 2.0\cdot10^{-2} | Helped |
| increase entropy parameter eps from 0 to 10 (ablation) | ENOT · CelebA super-resolution | LPIPS variability: 0 -> 3.4\cdot10^{-2} -> 5.1\cdot10^{-2} | Helped |
| increase entropy parameter eps from 0 to 10 (ablation) | ENOT · CelebA super-resolution | FID: 3.78 -> 7.63 -> 14.8 | Hurt |
| increase entropy parameter eps from 0 to 10 (ablation) | ENOT · Colored MNIST 2->3 | FID: 6.0 -> 6.28 -> 6.9 | Hurt |
| use ENOT instead of SCONES | unpaired super-resolution · CelebA 64x64 | FID: 18.88 -> 14.8 | Helped |
| use ENOT instead of DiffSB | Colored MNIST 2->3 · Colored MNIST | FID: 93 -> 6.28 | Helped |
| use ENOT instead of AugCycleGAN | unpaired super-resolution · CelebA 64x64 | FID: 15.2 -> 14.8 | Helped |
| use ENOT instead of FB-SDE-J | Gaussian EOT benchmark, dim 128 · EOT plan recovery | BW2^2-UVP: 0.58 -> 0.29 | Helped |
| use ENOT instead of FB-SDE-J | Gaussian EOT benchmark, dim 128 · learned marginal vs target | BW2^2-UVP: 0.39 -> 0.50 | Hurt |
| use ENOT instead of MLE-SB | Gaussian EOT benchmark, dim 128 · EOT plan recovery | BW2^2-UVP: 1.8 -> 0.29 | Helped |
Closest papers by meaning
- Progressive Entropic Optimal Transport Solvers (NeurIPS 2024 · similarity 0.91)
- Energy-guided Entropic Neural Optimal Transport (ICLR 2024 · similarity 0.89)
- Expectile Regularization for Fast and Accurate Training of Neural Optimal Transport (NeurIPS 2024 · similarity 0.87)
- Safe and Sparse Newton Method for Entropic-Regularized Optimal Transport (NeurIPS 2024 · similarity 0.87)
- Decreasing Entropic Regularization Averaged Gradient for Semi-Discrete Optimal Transport (NeurIPS 2025 · similarity 0.86)
- Transport meets Variational Inference: Controlled Monte Carlo Diffusions (ICLR 2024 · similarity 0.86)
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.