Research map / Optimization / Nonconvex and smooth optimization
Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile
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 an extra-gradient step to RMSprop | RMSprop GAN training · 16-Gaussian GMM | target-distribution learning | Helped |
| Add an extra-gradient step to RMSprop | RMSprop GAN training · 16-Gaussian GMM | oscillatory behavior | Helped |
| Add an extra-gradient step to Adam | Adam GAN training · 16-Gaussian GMM | target-distribution learning | Helped |
| Add an extra-gradient step to Adam | Adam GAN training · 16-Gaussian GMM | oscillatory behavior | Helped |
| Add an extra-gradient step to Adam | Wasserstein GAN · CelebA | inception score | Helped |
| Add an extra-gradient step to Adam | Wasserstein GAN · CelebA | Fréchet distance | Helped |
| Add an extra-gradient step to Adam | Wasserstein GAN · CIFAR-10 | inception score | Helped |
| Add an extra-gradient step to Adam | Wasserstein GAN · CIFAR-10 | Fréchet distance | Helped |
| Use vanilla mirror descent instead of optimistic mirror descent | Mirror descent in null-coherent problems · Coherent saddle-point problems | convergence to a solution | Hurt |
Closest papers by meaning
- Training GANs with Optimism (ICLR 2018 · similarity 0.90)
- A Variational Inequality Perspective on Generative Adversarial Networks (ICLR 2019 · similarity 0.89)
- Mirror Descent Beyond Euclidean Stability: An Exponential Separation in Initialization Sensitivity (NeurIPS 2026 · similarity 0.89)
- Never Saddle: Reparameterized Steepest Descent as Mirror Flow (ICLR 2026 · similarity 0.89)
- Towards Better Understanding of Adaptive Gradient Algorithms in Generative Adversarial Nets (ICLR 2020 · similarity 0.88)
- TRAINING GENERATIVE ADVERSARIAL NETWORKS VIA PRIMAL-DUAL SUBGRADIENT METHODS: A LAGRANGIAN PERSPECTIVE ON GAN (ICLR 2018 · 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.