Research map / Optimization / Nonconvex and smooth optimization
On the Last-iterate Convergence in Time-varying Zero-sum Games: Extra Gradient Succeeds where Optimism Fails
Yi Feng, Hu Fu, Qun Hu, Ping Li, Ioannis Panageas, bo peng, Xiao Wang
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 extra-gradient instead of OGDA in periodic zero-sum games | Extra-gradient · Periodic bilinear zero-sum games | last-iterate convergence to equilibrium | Improved |
| Use OGDA in the period-2 game with alternating payoff matrices | OGDA · Period-2 game with A_t = [1,-1] for odd t and [-1,1] for even t | distance to equilibrium | Hurt |
| Use negative momentum in the period-2 game with alternating payoff matrices | Negative momentum · Period-2 game with A_t = [1,-1] for odd t and [-1,1] for even t | distance to equilibrium | Hurt |
| Use EG, OGDA, or negative momentum with bounded accumulated perturbations | EG, OGDA, and negative momentum · Convergent perturbed bilinear zero-sum games satisfying BAP | distance to equilibrium of the stable game | Improved |
| Use EG without requiring bounded accumulated perturbations | Extra-gradient · Convergent perturbed bilinear zero-sum games | distance to equilibrium of the stable game | Improved |
| Use slower-decaying perturbations | EG, OGDA, and negative momentum · Convergent perturbed games satisfying BAP | convergence rate | Hurt |
| Increase perturbation magnitude | Extra-gradient · Convergent perturbed games without BAP | convergence rate | Hurt |
Closest papers by meaning
- Fast Last-Iterate Convergence of Learning in Games Requires Forgetful Algorithms (NeurIPS 2024 · similarity 0.91)
- Last-Iterate Convergence Properties of Regret-Matching Algorithms in Games (ICLR 2025 · similarity 0.90)
- Near-Optimal Last-Iterate Convergence for Zero-Sum Games with Bandit Feedback and Opponent Actions (NeurIPS 2026 · similarity 0.90)
- From Average-Iterate to Last-Iterate Convergence in Games: A Reduction and Its Applications (NeurIPS 2025 · similarity 0.89)
- Last Iterate Convergence in Monotone Mean Field Games (NeurIPS 2025 · similarity 0.88)
- Last-iterate Convergence in Regularized Graphon Mean Field Game (AAAI 2025 · 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.
