0 alternative definitions
An unsupervised learning technique used in reinforcement learning in which an agent learns to predict the value of a variable at the end of a sequence of states.
