Multi-Agent Training beyond Zero-Sum with Correlated Equilibrium Meta-Solvers
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2021
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914758485082112 |
|---|---|
| author | Marris, Luke Muller, Paul Lanctot, Marc Tuyls, Karl Graepel, Thore |
| author_facet | Marris, Luke Muller, Paul Lanctot, Marc Tuyls, Karl Graepel, Thore |
| contents | Two-player, constant-sum games are well studied in the literature, but there has been limited progress outside of this setting. We propose Joint Policy-Space Response Oracles (JPSRO), an algorithm for training agents in n-player, general-sum extensive form games, which provably converges to an equilibrium. We further suggest correlated equilibria (CE) as promising meta-solvers, and propose a novel solution concept Maximum Gini Correlated Equilibrium (MGCE), a principled and computationally efficient family of solutions for solving the correlated equilibrium selection problem. We conduct several experiments using CE meta-solvers for JPSRO and demonstrate convergence on n-player, general-sum games. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2106_09435 |
| institution | arXiv |
| publishDate | 2021 |
| record_format | arxiv |
| spellingShingle | Multi-Agent Training beyond Zero-Sum with Correlated Equilibrium Meta-Solvers Marris, Luke Muller, Paul Lanctot, Marc Tuyls, Karl Graepel, Thore Multiagent Systems Artificial Intelligence Computer Science and Game Theory Machine Learning Two-player, constant-sum games are well studied in the literature, but there has been limited progress outside of this setting. We propose Joint Policy-Space Response Oracles (JPSRO), an algorithm for training agents in n-player, general-sum extensive form games, which provably converges to an equilibrium. We further suggest correlated equilibria (CE) as promising meta-solvers, and propose a novel solution concept Maximum Gini Correlated Equilibrium (MGCE), a principled and computationally efficient family of solutions for solving the correlated equilibrium selection problem. We conduct several experiments using CE meta-solvers for JPSRO and demonstrate convergence on n-player, general-sum games. |
| title | Multi-Agent Training beyond Zero-Sum with Correlated Equilibrium Meta-Solvers |
| topic | Multiagent Systems Artificial Intelligence Computer Science and Game Theory Machine Learning |
| url | https://arxiv.org/abs/2106.09435 |