Research map / Optimization / Optimizers
Explore Aggressively, Update Conservatively: Stochastic Extragradient Methods with Variable Stepsize Scaling
Yu-Guan Hsieh, Franck Iutzeler, Jérôme Malick, Panayotis Mertikopoulos
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 a more aggressive exploration stepsize than the update stepsize (double stepsize) in stochastic extragradient (ablation) | stochastic extragradient (DSEG) · stochastic bilinear game | convergence of last iterate | Improved |
| use a more aggressive exploration stepsize than the update stepsize (double stepsize) in stochastic extragradient (ablation) | stochastic extragradient (DSEG) · non convex-concave linear quadratic Gaussian GAN | convergence of last iterate | Improved |
| use an aggressive exploration stepsize in addition to the update stepsize (ablation) | stochastic extragradient (DSEG) · strongly convex-concave game | convergence speed | No clear effect |
| use DSEG instead of anchored gradient descent | stochastic bilinear saddle-point example · stochastic bilinear game | convergence speed | Improved |
| use stochastic Hamiltonian gradient descent instead of DSEG | stochastic bilinear saddle-point example · stochastic bilinear game | early convergence speed | Mixed |
| use vanilla stochastic extragradient with any stepsize sequence | stochastic extragradient · bilinear min-max with noisy feedback | last-iterate convergence | Hurt |
Closest papers by meaning
- A Second look at Exponential and Cosine Step Sizes: Simplicity, Adaptivity, and Performance (ICML 2021 · similarity 0.89)
- Accelerated last-iterate convergence of Extragradient via power-law stepsizes (NeurIPS 2026 · similarity 0.89)
- Provable and Practical Online Learning Rate Adaptation with Hypergradient Descent (ICML 2025 · similarity 0.87)
- Nonasymptotic Analysis of Stochastic Gradient Descent with the Richardson–Romberg Extrapolation (ICLR 2025 · similarity 0.87)
- Double-Step Alternating Extragradient with Increasing Timescale Separation for Finding Local Minimax Points: Provable Improvements (ICML 2024 · similarity 0.86)
- Large Stepsizes Accelerate Gradient Descent for Regularized Logistic Regression (NeurIPS 2025 · 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.
