Consider the following grid world, in which an agent can explore the environment until it finds the Goal $($G$).$ In this problem, you will update the estimates of the Q function based on experiences of the agent. In this environment, all actions in all squares result in a zero reward, except the actions that result in entering the goal square and the actions that result entering the danger square X that result in a punishment, i$.$e$.$ negative reward. The rewards r$($s$,$ a$)$ of each action a in state s was shown in the below figure. Assume that the initial estimate Q$($s$,$a$)$ is zero for all state and action pairs.