Almost optimal algorithms for two-player zero-sum linear mixture Markov games

MaRDI QIDQ7028780zbMATH OpenFDO

Authors Zixiang Chen, Dongruo Zhou, Quanquan Gu

Publication date 11 February 2025

Full work available at URL https://proceedings.mlr.press/v167/chen22d.html

zbMATH Keywords

linear function approximationMarkov gamesreinforcement learning

Mathematics Subject Classification ID

Analysis of algorithms and problem complexity (68Q25) Learning and adaptive systems in artificial intelligence (68T05) Stochastic games, stochastic differential games (91A15)

Cites work

• 10.1162/1532443041827880
• Correlated Equilibrium as an Expression of Bayesian Rationality
• Finite-time bounds for fitted value iteration
• Near-optimal regret bounds for reinforcement learning
• Near-optimal reinforcement learning in polynomial time
• Stochastic Games
• Strategically zero-sum games: The class of games whose completely mixed equilibria cannot be improved upon
• Strategy iteration is strongly polynomial for 2-player turn-based stochastic games with a constant discount factor
• Superhuman AI for multiplayer poker
• Weighted sums of certain dependent random variables

Cited in

(2)

• Imprecise multi-armed bandits: representing irreducible uncertainty as a zero-sum game
• Provably efficient information-directed sampling algorithms for multi-agent reinforcement learning

This page was built for publication: Almost optimal algorithms for two-player zero-sum linear mixture Markov games

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7028780)