a) [7 points] Consider the following grid world in which you…

a) [7 points] Consider the following grid world in which you will implement TD learning and Q-learning techniques to find the values of these states. .  Suppose that we have the following observed transitions: (A, East, C, 3), (C, South, B, 4), (C, East, G, 4), (C, East, E, 3), (E, North, D, 3), (E, North, F, 4), (E, North, H, 6) The initial value of each state is 0. Assume that γ = 1 and α = 0.5. • What are the learned values from TD learning after all observations? • What are the learned Q-values from Q-learning after all observations? Show your procedure. Both solution and procedure will count towards the grade of this question. b) [3 points] Explain the difference between active and passive reinforcement learning.

Linear Models Suppose you want to learn a function that pred…

Linear Models Suppose you want to learn a function that predicts the season batting average of a baseball player, given the information available at the start of the season. (This is a simplified version of the premise of the movie Moneyball.)   (a) Explain how you would formalize this as a learning problem.  (b) Is it reasonable to model this relationship using a linear model? Why or why not?  (c) What machine learning techniques would you use? (d) Give a short explanation of how you might learn the model using an iterative algorithm