On Stochastic Optimal Control and Reinforcement Learning by Approximate Inference.
Konrad RawlikMarc ToussaintSethu VijayakumarPublished in: Robotics: Science and Systems (2012)
Keyphrases
- approximate inference
- reinforcement learning
- graphical models
- belief propagation
- probabilistic inference
- message passing
- exact inference
- gaussian process
- parameter estimation
- variational methods
- bayesian networks
- factor graphs
- loopy belief propagation
- latent variables
- dynamic bayesian networks
- conditional random fields
- expectation propagation
- state space
- free energy
- machine learning
- probabilistic model
- dynamic programming
- random variables
- structured prediction
- learning algorithm
- regression model
- supervised learning
- probability distribution