A Characterization of Meaningful Schedulers for Continuous-Time Markov Decision Processes.
Nicolás WolovickSven JohrPublished in: FORMATS (2006)
Keyphrases
- markov decision processes
- state space
- dynamic programming
- finite state
- optimal policy
- reinforcement learning
- markov chain
- reachability analysis
- stationary policies
- reinforcement learning algorithms
- policy iteration
- model based reinforcement learning
- markov decision process
- heuristic search
- transition matrices
- planning under uncertainty
- decision processes
- factored mdps
- average cost
- partially observable
- dynamical systems
- optimal control
- decision theoretic planning
- action space
- infinite horizon
- risk sensitive
- semi markov decision processes
- real time dynamic programming
- average reward
- grid computing
- planning problems
- discounted reward
- monte carlo
- probability distribution