Model-free finite-horizon optimal control of discrete-time two-player zero-sum games.
Wei WangXin ChenJianhua DuPublished in: Int. J. Syst. Sci. (2023)
Keyphrases
- optimal control
- model free
- finite horizon
- infinite horizon
- reinforcement learning algorithms
- optimal control problems
- reinforcement learning
- imperfect information
- optimal policy
- optimal strategy
- markov decision processes
- linear quadratic
- average cost
- finite state
- function approximation
- inventory control
- dynamic programming
- decision problems
- markov decision process
- state space
- single product
- production planning
- temporal difference
- policy iteration
- stochastic demand
- multi agent
- single agent
- impedance control
- average reward
- control law
- stochastic games
- periodic review
- multistage
- markov chain
- brownian motion
- machine learning
- learning algorithm
- lost sales
- fixed cost
- control policies
- reward function
- game playing
- action space
- long run
- markov decision problems
- nash equilibrium
- non stationary