Convergence of controlled models and finite-state approximation for discounted continuous-time Markov decision processes with constraints.
Xianping GuoWenzhao ZhangPublished in: Eur. J. Oper. Res. (2014)
Keyphrases
- markov decision processes
- finite state
- stationary policies
- continuous time bayesian networks
- transition matrices
- state space
- optimal policy
- policy iteration algorithm
- markov chain
- average cost
- finite state transducers
- reinforcement learning
- dynamic programming
- policy iteration
- decision theoretic planning
- reinforcement learning algorithms
- average reward
- action sets
- state and action spaces
- partially observable
- partially observable markov decision processes
- continuous state
- action space
- markov decision process
- infinite horizon
- dynamical systems
- random fields
- finite horizon
- heuristic search
- probabilistic model
- linear program