Inferring Smooth Control: Monte Carlo Posterior Policy Iteration with Gaussian Processes.
Joe WatsonJan PetersPublished in: CoRL (2022)
Keyphrases
- monte carlo
- gaussian processes
- gaussian process
- policy evaluation
- policy iteration
- temporal difference
- markov decision processes
- optimal control
- markov chain
- model free
- markov chain monte carlo
- hyperparameters
- importance sampling
- reinforcement learning
- particle filter
- finite state
- regression model
- fixed point
- approximate inference
- latent variables
- optimal policy
- model selection
- closed form
- multi task
- higher order
- least squares
- semi supervised
- probability distribution
- variance reduction
- dynamic programming