Efficient Solutions to Factored MDPs with Imprecise Transition Probabilities.
Karina Valdivia DelgadoScott SannerLeliane Nunes de BarrosFábio Gagliardi CozmanPublished in: ICAPS (2009)
Keyphrases
- efficient solutions
- transition probabilities
- factored mdps
- markov decision problems
- markov chain
- state space
- random walk
- context specific
- markov models
- markov decision processes
- reward function
- optimal solution
- approximate dynamic programming
- heuristic methods
- stochastic processes
- basis functions
- link structure
- reinforcement learning
- markov decision process
- hidden variables
- maximum entropy
- learning algorithm
- directed graph
- search space
- search algorithm
- objective function
- image segmentation