Safe Exploration in Finite Markov Decision Processes with Gaussian Processes.
Matteo TurchettaFelix BerkenkampAndreas KrausePublished in: CoRR (2016)
Keyphrases
- gaussian processes
- markov decision processes
- state and action spaces
- model based reinforcement learning
- gaussian process
- interval estimation
- optimal policy
- gaussian process regression
- state space
- reinforcement learning
- decision theoretic planning
- transition matrices
- dynamic programming
- policy iteration
- action space
- hyperparameters
- average cost
- multi task
- markov decision process
- finite number
- cross validation
- model selection
- pairwise
- search algorithm