Real-time dynamic programming for Markov decision processes with imprecise probabilities.
Karina Valdivia DelgadoLeliane N. de BarrosDaniel B. DiasScott SannerPublished in: Artif. Intell. (2016)
Keyphrases
- real time dynamic programming
- markov decision processes
- imprecise probabilities
- bayesian networks
- information processing
- finite state
- state space
- reinforcement learning
- probability theory
- belief functions
- optimal policy
- dynamic programming
- planning under uncertainty
- reinforcement learning algorithms
- policy iteration
- markov decision process
- average cost
- decision processes
- markov decision problems
- infinite horizon
- action space
- probability distribution
- decision making