DESPOT: Online POMDP Planning with Regularization.
Nan YeAdhiraj SomaniDavid HsuWee Sun LeePublished in: J. Artif. Intell. Res. (2017)
Keyphrases
- belief space
- planning problems
- partially observable markov decision processes
- online learning
- partially observable
- ai planning
- real time
- stochastic domains
- reinforcement learning
- dynamical systems
- machine learning
- partially observable markov decision process
- partial observability
- belief state
- regularization parameter
- decision problems
- state space
- multi agent
- planning under uncertainty
- domain independent
- point based value iteration
- partially observable stochastic domains