Planning with Continuous Actions in Partially Observable Environments.
Matthijs T. J. SpaanNikos A. VlassisPublished in: ICRA (2005)
Keyphrases
- partially observable
- partially observable environments
- decision problems
- reinforcement learning
- partial observability
- markov decision processes
- state space
- dynamical systems
- partially observable markov decision processes
- action selection
- inverse reinforcement learning
- action space
- reward function
- partially observable markov decision process
- decision theoretic
- goal directed
- infinite horizon
- planning domains
- planning problems
- belief state
- heuristic search
- initial state
- machine learning
- computational complexity
- optimal policy
- multi agent systems
- reinforcement learning algorithms
- dynamic programming