paper-with-me

Papers

Learning in two-player zero-sum partially observable Markov games with perfect recall

2021-12-01 · NeurIPS 2021 12 · Tadashi Kozuno, Pierre Ménard, Remi Munos, Michal Valko

We study the problem of learning a Nash equilibrium (NE) in an extensive game with imperfect information (EGII) through self-play. Precisely, we focus on two-player, zero-sum, episodic, tabular EGII under the \textit{perfect-recall} assumption where the only feedback is realizations of the game (bandit feedback). In particular the \textit{dynamics of the EGII is not known}---we can only access it by sampling or interacting with a game simulator. For this learning setting, we provide the Implicit Exploration Online Mirror Descent (IXOMD) algorithm. It is a model-free algorithm with a high-probability bound on convergence rate to the NE of order $1/\sqrt{T}$ where~$T$ is the number of played games. Moreover IXOMD is computationally efficient as it needs to perform the updates only along the sampled trajectory.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Model-Free Learning for Two-Player Zero-Sum Partially Observable Markov Games with Perfect Recall

2021-06-11 · Tadashi Kozuno, Pierre Ménard, Rémi Munos, Michal Valko

We study the problem of learning a Nash equilibrium (NE) in an imperfect information game (IIG) through self-play. Precisely, we focus on two-player, zero-sum, episodic, tabular IIG under the perfect-recall assumption wh…

HSVI can solve zero-sum Partially Observable Stochastic Games

2022-10-26 · Aurélien Delage, Olivier Buffet, Jilles S. Dibangoye, Abdallah Saffidine

State-of-the-art methods for solving 2-player zero-sum imperfect information games rely on linear programming or regret minimization, though not on dynamic programming (DP) or heuristic search (HS), while the latter are …

Decision MakingHeuristic SearchOpen-Ended Question AnsweringSequential Decision Making

On Bellman's Optimality Principle for zs-POSGs

2020-06-29 · Olivier Buffet, Jilles Dibangoye, Aurélien Delage, Abdallah Saffidine 외

Many non-trivial sequential decision-making problems are efficiently solved by relying on Bellman's optimality principle, i.e., exploiting the fact that sub-problems are nested recursively within the original problem. He…

Decision MakingHeuristic SearchSequential Decision Making

Independent Learning of Nash Equilibria in Partially Observable Markov Potential Games with Decoupled Dynamics

2026-05-07 · Philip Jordan, Maryam Kamgarpour arxiv

We study Nash equilibrium learning in partially observable Markov games (POMGs), a multi-agent reinforcement learning framework in which agents cannot fully observe the underlying state. Prior work in this setting relies…

Multi-agent Reinforcement Learning

Solving Hierarchical Information-Sharing Dec-POMDPs: An Extensive-Form Game Approach

2024-02-05 · Johan Peralez, Aurélien Delage, Olivier Buffet, Jilles S. Dibangoye

A recent theory shows that a multi-player decentralized partially observable Markov decision process can be transformed into an equivalent single-player game, enabling the application of \citeauthor{bellman}'s principle …

FormManagement