Research map / Image generation / GANs
TRAINING GENERATIVE ADVERSARIAL NETWORKS VIA PRIMAL-DUAL SUBGRADIENT METHODS: A LAGRANGIAN PERSPECTIVE ON GAN
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 |
|---|---|---|---|
| add the primal-dual generator objective | GAN · MNIST | Inception score: 8.8 -> 9.8 | Helped |
| add the primal-dual generator objective | GAN · CIFAR-10 | Inception score: $3.82 \pm 0.06$ -> $4.53 \pm 0.04$ | Helped |
| add the primal-dual generator objective | GAN · CIFAR-10 | Inception score: $6.40 \pm 0.05$ -> $4.53 \pm 0.04$ | Hurt |
| add the second generator regularization term (ablation) | GAN · 1000-class MNIST | Modes generated: 526 -> 827 | Helped |
| add the second generator regularization term (ablation) | GAN · 1000-class MNIST | Inception score: 87.15 -> 155.6 | Helped |
| add the primal-dual generator objective | GAN · toy example with data at x=1 and initial generated samples at x=-3.0 | mode collapse | Helped |
| add the primal-dual generator objective | GAN · 2D mixture of 8 Gaussians | mode coverage | Helped |
Closest papers by meaning
- A Convex Duality Framework for GANs (NeurIPS 2018 · similarity 0.93)
- Dualing GANs (NeurIPS 2017 · similarity 0.92)
- Hidden Convexity of Wasserstein GANs: Interpretable Generative Models with Closed-Form Solutions (ICLR 2022 · similarity 0.92)
- SGAN: An Alternative Training of Generative Adversarial Networks (CVPR 2018 · similarity 0.91)
- A Variational Inequality Perspective on Generative Adversarial Networks (ICLR 2019 · similarity 0.91)
- Dual Discriminator Generative Adversarial Nets (NeurIPS 2017 · similarity 0.91)
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.