Research map / Learning theory and algorithms / Submodular and game algorithms
Learning to Correlate in Multi-Player General-Sum Sequential 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 deterministic joint-distribution reconstruction instead of sampling normal-form plans | CFR-Jr versus CFR-S · standard testbeds of multi-player general-sum sequential games | runtime to approximate CCE | Helped |
| Use CFR-Jr rather than CFR-S | multi-player general-sum sequential games · general-sum test instances | social welfare approximation ratio | No clear effect |
| Use CFR-Jr rather than vanilla CFR | multi-player general-sum sequential games · G2-4-DA | CCE approximation | Helped |
| Use CFR-Jr rather than vanilla CFR | multi-player general-sum sequential games · G2-4-DA | social welfare approximation ratio | Helped |
| Track the empirical joint frequency of play instead of taking the product of players' average strategies | regret-minimizing strategies in general-sum games · two-player example game | CCE convergence | Helped |
| Use CFR-S sampling to retain joint normal-form plans | CFR in multi-player general-sum extensive-form games · multi-player general-sum extensive-form games | CCE convergence | Helped |
| Use CFR behavioral-strategy averages without joint-strategy tracking | CFR in multi-player general-sum games · general-sum games | CCE convergence | Hurt |
| Reconstruct normal-form strategies from behavioral strategies directly on the game tree | CFR-Jr · multi-player general-sum extensive-form games | CCE approximation | Helped |
| Seek a social-welfare-maximizing CCE in general sequential games with two players and chance, or three or more players | optimal CCE computation · general-sum extensive-form games | computational approximability | Hurt |
Closest papers by meaning
- Coarse Correlation in Extensive-Form Games (AAAI 2020 · similarity 0.91)
- Hindsight and Sequential Rationality of Correlated Play (AAAI 2021 · similarity 0.91)
- When Can We Learn General-Sum Markov Games with a Large Number of Players Sample-Efficiently? (ICLR 2022 · similarity 0.91)
- Mediator Interpretation and Faster Learning Algorithms for Linear Correlated Equilibria in General Sequential Games (ICLR 2024 · similarity 0.90)
- Learning Rationalizable Equilibria in Multiplayer Games (ICLR 2023 · similarity 0.90)
- The Complexity of Correlated Equilibria in Generalized Games (NeurIPS 2025 · similarity 0.89)
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.