Research map / Optimization / Nonconvex and smooth optimization
Efficient Mirror Descent Ascent Methods for Nonsmooth Minimax Problems
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 variance reduction (VR-SMDA) on top of SMDA (ablation) | stochastic mirror descent ascent · robust NN training, Fashion-MNIST and MNIST | loss vs epoch | Helped |
| mirror descent ascent with adaptive preconditioning (Adam-type H_t, G_t) instead of proximal GDA baselines | MDA/SMDA/VR-SMDA · fair classifier, Fashion-MNIST, MNIST, CIFAR-10 | loss | Helped |
| mirror methods vs PASGDA/PAGDA/MAPGDA | MDA/SMDA/VR-SMDA · robust NN training | loss | Helped |
| MDA vs PAGDA deterministic complexity | MDA · nonsmooth minimax theory | gradient complexity in kappa: O(κ2ϵ−2) -> O(√κϵ−2) | Helped |
| SMDA vs PASGDA stochastic complexity | SMDA · nonsmooth minimax theory | gradient complexity: O(κ3ϵ−4) vs O(κ3ϵ−4) | No clear effect |
Closest papers by meaning
- Stochastic Composite Mirror Descent: Optimal Bounds with High Probabilities (NeurIPS 2018 · similarity 0.90)
- Stochastic Gradient/Mirror Descent: Minimax Optimality and Implicit Regularization (ICLR 2019 · similarity 0.90)
- Stochastic Mirror Descent in Variationally Coherent Optimization Problems (NeurIPS 2017 · similarity 0.90)
- High-probability complexity bounds for stochastic non-convex minimax optimization (NeurIPS 2024 · similarity 0.87)
- Mirror Descent-Ascent for mean-field min-max problems (NeurIPS 2026 · similarity 0.87)
- Last-Iterate Convergence of Single-Loop Stochastic Methods for Constrained Convex-Concave Minimax Problems (NeurIPS 2026 · 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.