Research map / Learning theory and algorithms / Optimal transport
Discriminator 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 |
|---|---|---|---|
| apply latent space DOT after sampling | DCGAN WGAN-GP · CIFAR-10 | FID: 27.84 -> 24.14 | Helped |
| apply latent space DOT after sampling | SNGAN(ns) · CIFAR-10 | FID: 20.74 -> 15.78 | Helped |
| apply latent space DOT after sampling | DCGAN WGAN-GP · STL-10 | FID: 49.94 -> 44.45 | Helped |
| apply conditional latent DOT (32 updates) | conditional SNGAN projection · ImageNet | FID: 43.34 -> 42.35 | Helped |
| use latent space DOT instead of target space DOT (ablation) | DOT · high-dimensional images | visual quality | Helped |
| use DOT instead of naive discriminator ascent (ablation) | target space transport · swissroll WGAN-GP | EMD: 0.160(22) -> 0.020(06) | Helped |
| use DOT instead of naive discriminator ascent (ablation) | target space transport · 25gaussians WGAN-GP | EMD: 0.065(11) -> 0.052(10) | Helped |
| use SNGAN(ns) discriminator in DOT versus none (ablation) | SAGAN(hi) DCGAN generator · CIFAR-10 | FID: 25.78 -> 21.45 | Helped |
| use WGAN-gp discriminator in DOT versus none (ablation) | SAGAN(hi) DCGAN generator · CIFAR-10 | FID: 25.78 -> 24.47 | Helped |
| skip gradient projection onto sphere for Gaussian prior (ablation) | latent DOT · CIFAR-10 | IS/FID | Hurt |
| more DOT updates (4 to 32) (ablation) | conditional DOT · ImageNet | FID: 43.34 -> 42.35 | Helped |
| increase step size epsilon (ablation) | latent DOT · CIFAR-10 | IS/FID | Mixed |
Closest papers by meaning
- Improving GANs Using Optimal Transport (ICLR 2018 · similarity 0.91)
- On the Convergence and Robustness of Training GANs with Regularized Optimal Transport (NeurIPS 2018 · similarity 0.91)
- Parallel Optimal Transport GAN (CVPR 2019 · similarity 0.90)
- Analyzing and Improving Optimal-Transport-based Adversarial Networks (ICLR 2024 · similarity 0.89)
- A Statistical Learning Perspective on Semi-dual Adversarial Neural Optimal Transport Solvers (ICLR 2026 · similarity 0.88)
- MinMax Methods for Optimal Transport and Beyond: Regularization, Approximation and Numerics (NeurIPS 2020 · 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.