Adaptive policy for two finite Markov chains zero-sum stochastic game with unknown transition matrices and average payoffs

DOI10.1016/S0005-1098(01)00050-4zbMath0989.93097OpenAlexW2117196412MaRDI QIDQ5939326

No author found.

Publication date: 5 August 2002

Published in: Automatica (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/s0005-1098(01)00050-4

zbMATH Keywords

saddle-pointadaptive policyaverage payoffscompetitive Markovian decision processesfinite stochastic gamesirreducible Markov chainslearning control policiesnormalization procedureregularized Lagrangian functionzero-sum game

Mathematics Subject Classification ID

Least squares and related methods for stochastic control systems (93E24) Stochastic games, stochastic differential games (91A15)

Related Items (3)

Average optimal strategies for zero-sum Markov games with poorly known payoff function on one side ⋮ An approach problem for a discrete system with random perturbations ⋮ Dr. Alexander Semionovich Poznyak Gorbatch: Biography

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Adaptive control of constrained Markov chains: Criteria and policies
• Stochastic zero-sum game of two automata
• Adaptive control of constrained finite Markov chains
• Learning automata and stochastic optimization
• Markov games with incomplete information
• Stationary \(\varepsilon\)-optimal strategies in stochastic games
• Finite state Markovian decision processes
• Conflict controlled processes with discontinuous trajectories
• Finite Continuous Time Markov Chains
• Algorithms for stochastic games ? A survey
• MARKOV STRATEGIES ARE BETTER THAN STATIONARY STRATEGIES
• Noncooperative Stochastic Games
• Stochastic games

This page was built for publication: Adaptive policy for two finite Markov chains zero-sum stochastic game with unknown transition matrices and average payoffs