Finite optimal control for time-bounded reachability in CTMDPs and continuous-time Markov games.
Markus N. RabeSven SchewePublished in: Acta Informatica (2011)
Keyphrases
- optimal control
- markov games
- control problems
- reinforcement learning
- state space
- markov decision processes
- dynamic programming
- optimal control problems
- reinforcement learning algorithms
- markov decision process
- control strategy
- infinite horizon
- multiagent reinforcement learning
- control law
- finite number
- neural network
- continuous functions
- optimal policy
- average cost
- real time
- function approximation
- brownian motion
- machine learning