mlpy.learners.online.rl.RLDTLearner.learn¶
-
RLDTLearner.
learn
(experience=None)[source]¶ Learn a policy from the experience.
A policy is learned from the experience by building the MDP model.
Parameters: experience : Experience
The actor’s current experience consisting of previous state, the action performed in that state, the current state, and the reward awarded.