Preference-Based Policy Iteration: Leveraging Preference Learning for Reinforcement Learning.
Weiwei ChengJohannes FürnkranzEyke HüllermeierSang-Hyeun ParkPublished in: ECML/PKDD (1) (2011)
Keyphrases
- policy iteration
- preference learning
- reinforcement learning
- markov decision processes
- model free
- gaussian processes
- optimal policy
- ordinal regression
- temporal difference
- pairwise comparison
- policy evaluation
- markov decision process
- function approximation
- active learning
- fixed point
- recommender systems
- ranking functions
- state space
- average reward
- supervised learning
- finite state
- preference relations
- least squares
- multi agent
- learning algorithm
- learning problems
- optimal control
- convergence rate
- linear programming
- learning process
- support vector
- evaluation function
- machine learning
- transfer learning
- em algorithm
- sufficient conditions
- policy search