Online POMDP Planning with Anytime Deterministic Guarantees.
Moran BarenboimVadim IndelmanPublished in: CoRR (2023)
Keyphrases
- stochastic domains
- planning problems
- partially observable markov decision processes
- belief space
- fully observable
- reinforcement learning
- partially observable
- state space
- online learning
- partially observable markov decision process
- belief state
- dynamical systems
- deterministic domains
- initial state
- heuristic search
- hidden state
- real time
- learning algorithm
- optimal planning
- finite state
- optimal policy
- initially unknown
- dynamic environments
- multi agent
- knowledge base
- predictive state representations
- partially observable stochastic domains