Model-based Bayesian Reinforcement Learning in Partially Observable Domains.
Pascal PoupartNikos A. VlassisPublished in: ISAIM (2008)
Keyphrases
- partially observable domains
- bayesian reinforcement learning
- partially observable markov decision processes
- reinforcement learning
- optimal policy
- monte carlo tree search
- finite state
- dynamical systems
- model free
- partially observable
- decision problems
- dynamic programming
- markov decision processes
- state space
- action models
- inverse reinforcement learning
- temporal reasoning
- model checking
- infinite horizon
- approximate solutions
- monte carlo
- sensing actions
- computational complexity