What is Decidable about Partially Observable Markov Decision Processes with ω-Regular Objectives.
Krishnendu ChatterjeeMartin ChmelikMathieu TracolPublished in: CoRR (2013)
Keyphrases
- partially observable markov decision processes
- finite state
- reinforcement learning
- dynamical systems
- planning under uncertainty
- belief state
- dynamic programming
- optimal policy
- continuous state
- decision problems
- partially observable stochastic games
- belief space
- markov decision processes
- stochastic domains
- multi agent
- partially observable markov
- state space
- partially observable
- planning problems
- partial observability
- approximate solutions
- partially observable domains
- dec pomdps
- predictive state representations
- sequential decision making problems
- markov chain
- first order logic
- long run
- utility function
- model checking
- partially observable markov decision process
- point based value iteration