Automatic Recovery Using Bounded Partially Observable Markov Decision Processes.
Kaustubh R. JoshiWilliam H. SandersMatti A. HiltunenRichard D. SchlichtingPublished in: DSN (2006)
Keyphrases
- partially observable markov decision processes
- finite state
- belief space
- reinforcement learning
- stochastic domains
- dynamical systems
- belief state
- partially observable stochastic games
- dynamic programming
- decision problems
- partial observability
- planning under uncertainty
- fully observable
- continuous state
- partially observable markov
- dec pomdps
- markov decision processes
- linear programming
- decision making
- partially observable
- multi agent
- partially observable domains
- sequential decision making problems
- learning algorithm