Mean-Variance Criteria for Finite Continuous-Time Markov Decision Processes.
Xianping GuoXinYuan SongPublished in: IEEE Trans. Autom. Control. (2009)
Keyphrases
- markov decision processes
- state space
- state and action spaces
- stationary policies
- optimal policy
- finite state
- dynamic programming
- reinforcement learning
- markov chain
- policy iteration
- transition matrices
- action sets
- markov decision process
- risk sensitive
- action space
- optimal control
- reinforcement learning algorithms
- average cost
- reachability analysis
- factored mdps
- decision processes
- decision theoretic planning
- finite horizon
- utility function
- partially observable
- dynamical systems
- average reward
- planning under uncertainty
- infinite horizon
- model based reinforcement learning
- finite number
- heuristic search
- optimality criterion
- least squares
- markov decision problems
- stochastic processes
- probability distribution