Strong Uniform Value in Gambling Houses and Partially Observable Markov Decision Processes.
Xavier VenelBruno ZiliottoPublished in: SIAM J. Control. Optim. (2016)
Keyphrases
- partially observable markov decision processes
- finite state
- belief space
- reinforcement learning
- belief state
- dynamical systems
- planning under uncertainty
- optimal policy
- continuous state
- state space
- partially observable stochastic games
- dynamic programming
- decision problems
- partial observability
- stochastic domains
- markov decision processes
- multi agent
- partially observable
- planning problems
- partially observable domains
- point based value iteration
- sequential decision making problems
- partially observable markov
- approximate solutions
- predictive state representations
- infinite horizon
- orders of magnitude
- decision trees
- learning algorithm