Empirical approximation in Markov games under unbounded payoff: discounted and average criteria.
Fernando Luque-VásquezJ. Adolfo Minjárez-SosaPublished in: Kybernetika (2017)
Keyphrases
- markov games
- markov decision processes
- markov decision process
- average cost
- multiagent reinforcement learning
- state space
- dynamic programming
- reinforcement learning algorithms
- optimal policy
- infinite horizon
- finite state
- stochastic games
- queueing networks
- finite horizon
- approximation methods
- upper bound
- search algorithm
- policy iteration
- average reward
- multiagent systems
- machine learning
- worst case
- cooperative
- reinforcement learning