Explicitly modeling generalization functionality lets MBTL to estimate the value of coaching on a whole new task. Reinforcement learning: A computer program interacts using a dynamic natural environment in which it have to accomplish a specific goal (which include driving a automobile or taking p