Consider the MDP below, representing a gynmastics robot on a balance beam. Like the Red Rocks, the robot often falls off the beam. Each grid square is a state and the available actions are right and left. States s1 and s5 represent different ends of the routine without falling, but moving right is apparently much more spectacular: R(s2, L, s1) 1 versus R(s, R, s5 10. Falling receives a negative reward R(s, L V R,G)-1 to the terminal ground state G. All other rewards are zero. +1 +10 ground Moving left or right results in a move left or right (respectively) with probability p. With probability 1 - p, the robot falls off the beam. Suppose y 1. Perform two iterations of value iteration. Show your work; e.g. Q:(82, L) 2. Find the range of values for p for which the best policy is to go left in state s2 Consider the MDP below, representing a gynmastics robot on a balance beam. Like the Red Rocks, the robot often falls off the beam. Each grid square is a state and the available actions are right and left. States s1 and s5 represent different ends of the routine without falling, but moving right is apparently much more spectacular: R(s2, L, s1) 1 versus R(s, R, s5 10. Falling receives a negative reward R(s, L V R,G)-1 to the terminal ground state G. All other rewards are zero. +1 +10 ground Moving left or right results in a move left or right (respectively) with probability p. With probability 1 - p, the robot falls off the beam. Suppose y 1. Perform two iterations of value iteration. Show your work; e.g. Q:(82, L) 2. Find the range of values for p for which the best policy is to go left in state s2