Approximate Planning for Factored POMDPs using Belief State Simplification
David A. McAllesterSatinder P. SinghPublished in: CoRR (2013)
Keyphrases
- belief state
- point based value iteration
- state space
- belief space
- partially observable markov decision processes
- dynamic bayesian networks
- partially observable
- partial observability
- planning problems
- belief revision
- partial knowledge
- stochastic domains
- approximation methods
- reactive planning
- heuristic search
- markov chain
- state variables
- planning under uncertainty
- partially observable markov decision process
- dynamic programming
- optimal policy
- reinforcement learning
- continuous state
- learning algorithm
- markov decision processes
- objective function
- approximate solutions
- decision problems
- dynamic environments