Partially Observable Markov Decision Processes incorporating epistemic uncertainties.
R. FaddoulWassim RaphaelA.-H. SoubraAlaa ChateauneufPublished in: Eur. J. Oper. Res. (2015)
Keyphrases
- partially observable markov decision processes
- finite state
- reinforcement learning
- belief space
- dynamical systems
- belief state
- planning under uncertainty
- decision problems
- optimal policy
- dynamic programming
- continuous state
- markov decision processes
- partially observable stochastic games
- stochastic domains
- partial observability
- partially observable domains
- planning problems
- state space
- sequential decision making problems
- partially observable
- multi agent
- partially observable markov decision process
- orders of magnitude
- approximate solutions
- machine learning
- partially observable markov
- initial state
- belief revision
- dec pomdps
- predictive state representations
- search space
- computational complexity