Research map / Learning theory and algorithms / Submodular and game algorithms
Meta-Learning in 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 |
|---|---|---|---|
| Meta-learn the OGD initialization from previous games | OGD · Libratus poker endgames, Endgame A and Endgame B | task-averaged Nash equilibrium gap: up to an order of magnitude | Helped |
| Initialize each game with the last iterate of the previous game instead of the vanilla OGD initialization (ablation) | OGD · Libratus poker endgames, sorted stacks | task-averaged Nash equilibrium gap | Helped |
| Initialize according to the meta-learning method instead of using the previous game's last iterate (ablation) | OGD · Libratus poker endgames, alternating and random stacks | task-averaged Nash equilibrium gap | Helped |
| Use the vanilla OGD initialization and solve each game separately rather than meta-learning initialization (ablation) | OGD · Libratus poker endgames, all task sequencing setups | task-averaged Nash equilibrium gap: up to an order of magnitude | Hurt |
Closest papers by meaning
- Learning Not to Regret (AAAI 2024 · similarity 0.91)
- Neural Auto-Curricula in Two-Player Zero-Sum Games (NeurIPS 2021 · similarity 0.90)
- The Equivalence of Dynamic and Strategic Stability under Regularized Learning in Games (NeurIPS 2023 · similarity 0.89)
- Securing Equal Share: A Principled Approach for Learning Multiplayer Symmetric Games (ICML 2025 · similarity 0.89)
- From Average-Iterate to Last-Iterate Convergence in Games: A Reduction and Its Applications (NeurIPS 2025 · similarity 0.88)
- No-regret Learning in Harmonic Games: Extrapolation in the Face of Conflicting Interests (NeurIPS 2024 · 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.