Multiple-Environment Markov Decision Processes.
Jean-François RaskinOcan SankurPublished in: FSTTCS (2014)
Keyphrases
- markov decision processes
- optimal policy
- decision processes
- finite state
- dynamic programming
- state space
- reinforcement learning
- transition matrices
- decision theoretic planning
- policy iteration
- finite horizon
- average cost
- planning under uncertainty
- risk sensitive
- reachability analysis
- model based reinforcement learning
- markov decision process
- action sets
- dynamic environments
- average reward
- interval estimation
- reinforcement learning algorithms
- decision diagrams
- state abstraction
- infinite horizon