Prioritizing Point-Based POMDP Solvers.
Guy ShaniRonen I. BrafmanSolomon Eyal ShimonyPublished in: ECML (2006)
Keyphrases
- point based value iteration
- belief state
- reinforcement learning
- belief space
- partially observable markov decision processes
- continuous state
- optimal policy
- dynamical systems
- sat solvers
- finite state
- state space
- eigenvalue problems
- partial observability
- partially observable markov decision process
- partially observable
- dec pomdps
- multi agent
- highly optimized
- numerically stable
- computationally hard problems
- requirements engineering
- markov decision processes
- np complete
- least squares
- mobile robot
- search space