Continuous-time Markov decision processes with nth-bias optimality criteria.
Junyu ZhangXi-Ren CaoPublished in: Autom. (2009)
Keyphrases
- markov decision processes
- optimality criteria
- state space
- finite state
- dynamic programming
- optimal policy
- reinforcement learning
- transition matrices
- markov chain
- policy iteration
- stationary policies
- optimal control
- planning under uncertainty
- reinforcement learning algorithms
- reachability analysis
- average cost
- decision theoretic planning
- average reward
- factored mdps
- action space
- markov decision process
- partially observable
- infinite horizon
- dynamical systems
- model based reinforcement learning
- heuristic search
- state and action spaces
- finite horizon
- variance reduction
- planning problems
- semi markov decision processes
- objective function
- model checking
- action sets
- evaluation function
- state abstraction
- risk sensitive
- reward function