Purely Epistemic Markov Decision Processes.
Régis SabbadinJérôme LangNasolo RavoanjanahryPublished in: AAAI (2007)
Keyphrases
- markov decision processes
- finite state
- reinforcement learning
- optimal policy
- policy iteration
- transition matrices
- state space
- finite horizon
- dynamic programming
- infinite horizon
- decision theoretic planning
- decision processes
- reachability analysis
- planning under uncertainty
- model based reinforcement learning
- reinforcement learning algorithms
- partially observable
- factored mdps
- risk sensitive
- machine learning
- average reward
- action space
- search algorithm
- markov decision process
- average cost
- reward function
- linear program