Anytime Point-Based Approximations for Large POMDPs
Joelle PineauGeoffrey J. GordonSebastian ThrunPublished in: CoRR (2011)
Keyphrases
- point based value iteration
- belief state
- approximation methods
- reinforcement learning
- partially observable markov decision processes
- belief space
- markov decision processes
- closed form
- state space
- continuous state
- neural network
- planning under uncertainty
- efficient computation
- distributed constraint optimization
- optimal control
- domain independent
- computational complexity
- decision trees