Percentile Queries in Multi-Dimensional Markov Decision Processes.
Mickael RandourJean-François RaskinOcan SankurPublished in: CoRR (2014)
Keyphrases
- markov decision processes
- multi dimensional
- range queries
- optimal policy
- finite state
- state space
- reinforcement learning
- policy iteration
- dynamic programming
- query processing
- decision theoretic planning
- reachability analysis
- transition matrices
- finite horizon
- planning under uncertainty
- model based reinforcement learning
- action sets
- decision processes
- action space
- infinite horizon
- average cost
- partially observable
- factored mdps
- state and action spaces
- markov decision process
- partially observable markov decision processes
- machine learning
- stochastic shortest path
- reinforcement learning algorithms
- risk sensitive
- stationary policies
- least squares
- interval estimation
- learning algorithm