Bayesian reinforcement learning in continuous POMDPs with gaussian processes.
Patrick DallaireCamille BesseStéphane RossBrahim Chaib-draaPublished in: IROS (2009)
Keyphrases
- gaussian processes
- bayesian reinforcement learning
- partially observable markov decision processes
- optimal policy
- reinforcement learning
- monte carlo tree search
- gaussian process
- finite state
- gaussian process regression
- belief state
- monte carlo
- dynamic programming
- covariance function
- decision problems
- markov decision processes
- hyperparameters
- dynamical systems
- state space
- multi task
- search algorithm
- gaussian process models
- bayesian inference
- random sampling
- machine learning
- semi supervised
- probabilistic model
- learning algorithm