A survey of point-based POMDP solvers.
Guy ShaniJoelle PineauRobert KaplowPublished in: Auton. Agents Multi Agent Syst. (2013)
Keyphrases
- point based value iteration
- belief space
- belief state
- reinforcement learning
- state space
- continuous state
- partially observable markov decision processes
- partially observable
- model free reinforcement learning
- highly optimized
- markov decision processes
- dynamical systems
- hidden state
- planning problems
- quantified boolean formulas
- optimal policy
- robot navigation
- knowledge base
- sat solvers
- constraint satisfaction
- sat solving
- numerically stable
- partially observable stochastic games