Forward Search Value Iteration for POMDPs.
Guy ShaniRonen I. BrafmanSolomon Eyal ShimonyPublished in: IJCAI (2007)
Keyphrases
- forward search
- state space
- partially observable markov decision processes
- belief state
- heuristic search
- markov decision processes
- partially observable
- reinforcement learning
- belief space
- partially observable markov
- optimal policy
- planning problems
- dynamic programming
- finite state
- continuous state
- state space search
- orders of magnitude
- planning under uncertainty
- dynamical systems
- partial observability
- markov decision process
- markov chain
- markov decision problems
- search space
- dec pomdps
- search algorithm
- initial state
- average reward
- action space
- action sequences
- infinite horizon
- decision problems
- multi agent
- average cost
- reward function
- optimal solution