Zero-sum infinite-horizon discounted piecewise deterministic Markov games.
Yonghui HuangZhaotong LianXianping GuoPublished in: Math. Methods Oper. Res. (2023)
Keyphrases
- markov games
- infinite horizon
- markov decision processes
- markov decision process
- optimal policy
- stochastic games
- finite horizon
- multiagent reinforcement learning
- stationary policies
- dynamic programming
- state space
- finite state
- optimal control
- long run
- partially observable
- average cost
- policy iteration
- reinforcement learning algorithms
- reinforcement learning
- action space
- lead time
- average reward
- initial state
- bayesian networks
- markov decision problems
- sufficient conditions
- probability distribution