Research map / Bandits and online learning / Online convex optimization and games
Last-iterate Convergence in Extensive-Form Games
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 optimistic mirror descent with a strongly convex, continuously differentiable regularizer | Zero-sum EFGs with perfect recall · Zero-sum EFGs with perfect recall | last-iterate convergence | Helped |
| Use the vanilla squared Euclidean regularizer in optimistic mirror descent | VOGDA · Zero-sum EFGs with perfect recall | last-iterate convergence rate: dist^2(z_t, Z*) ≤ 64 dist^2(z'_1, Z*)(1+C_1)^(-t) | Helped |
| Use the vanilla entropy regularizer in optimistic mirror descent | VOMWU · Zero-sum EFGs with perfect recall and a unique Nash equilibrium | last-iterate convergence rate: C_2/t | Helped |
| Use the dilated entropy regularizer in optimistic mirror descent | DOMWU · Zero-sum EFGs with perfect recall and a unique Nash equilibrium | last-iterate convergence rate: C_3(1+C_4)^(-t) | Helped |
| Use optimistic global regret minimization instead of CFR+ | Kuhn poker · Kuhn poker | duality gap | Helped |
| Use optimistic global regret minimization instead of CFR+ | Pursuit-evasion · Pursuit-evasion | duality gap | Helped |
| Change CFR+ from simultaneous updates to alternating updates | CFR+ with regret matching+ · Leduc poker | duality gap | Helped |
| Use CFR with the initialization described in the theorem | CFR · Rock-paper-scissors | last-iterate convergence | Hurt |
| Use simultaneous updates for CFR, CFR+, or their optimistic versions | CFR-based algorithms · Rock-paper-scissors | last-iterate convergence | Hurt |
Closest papers by meaning
- Optimistic Regret Minimization for Extensive-Form Games via Dilated Distance-Generating Functions (NeurIPS 2019 · similarity 0.94)
- Last-Iterate Convergence Properties of Regret-Matching Algorithms in Games (ICLR 2025 · similarity 0.93)
- Efficient Last-Iterate Convergence in Solving Extensive-Form Games (NeurIPS 2025 · similarity 0.93)
- Fast Last-Iterate Convergence of Learning in Games Requires Forgetful Algorithms (NeurIPS 2024 · similarity 0.91)
- Optimism Without Regularization: Constant Regret in Zero-Sum Games (NeurIPS 2025 · similarity 0.90)
- Near-Optimal Last-Iterate Convergence for Zero-Sum Games with Bandit Feedback and Opponent Actions (NeurIPS 2026 · similarity 0.90)
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.