Monte Carlo Methods

L603 Monte Carlo Methods RENDER V3-2

## Quiz

To check your understanding of the video, please answer the question below.

Which of the following describes the Monte Carlo approach discussed in the video?

After each time step, the agent will select a different action.

For the first episode, the agent selects the first action at every time step. For the second episode, it selects a different action, and so on.

When the agent has a policy in mind, it follows the policy to collect a lot of episodes. Then, for each state, to figure out which action is best, the agent can look for which action tended to result in the most cumulative reward.

When the agent has a policy in mind, it follows the policy to collect a single episode. The agent uses the episode to tell if the policy is good or bad by looking at the cumulative reward that was received by the agent.

SOLUTION:

When the agent has a policy in mind, it follows the policy to collect a lot of episodes. Then, for each state, to figure out which action is best, the agent can look for which action tended to result in the most cumulative reward.