Homework (2/2) 2-a What are the optimal state-values and state-action-values for this environment? 2-b What is the optimal policy for this environment? 2-c Assume we introduce a discount factor of 0.95 into our value functions. Determine the new values of the state-value and state-action-value functions as well as the new optimal policy. Describe the effect of the discount factor on the optimal policy. 3. (2pts) we will formulate Tic-Tac-Toe as an environment in which we can train a reinforcement learning agent. You will play as X's, and your opponent will be O's. Two-player games such as Tic-Tac-Toe are often modeled using game theory, in which we try and predict the moves of our opponent as well. For simplicity, we ignore the modeling of the opponent moves and treat our opponent's actions as a source of randomness within the environment. Assume you always go first. What are the states and actions within the Tic-Tac-Toe reinforcement learning environment? How does the current state affect the actions you can take?