Online Markov Decision Processes.
Eyal Even-DarSham M. KakadeYishay MansourPublished in: Math. Oper. Res. (2009)
Keyphrases
- markov decision processes
- optimal policy
- finite state
- state space
- decision theoretic planning
- transition matrices
- policy iteration
- dynamic programming
- reinforcement learning
- risk sensitive
- reinforcement learning algorithms
- decision processes
- action sets
- finite horizon
- reachability analysis
- partially observable
- average reward
- model based reinforcement learning
- state and action spaces
- average cost
- planning under uncertainty
- reward function
- markov decision process
- action space
- discounted reward
- decision making
- factored mdps
- learning algorithm
- infinite horizon
- optimal solution