Login to unlock
Machine Learning Fundamentals · Phase 8: Reinforcement Learning · Agents, Rewards, and Q-Learning

Agents, Rewards, and Q-Learning — Checkpoint Quiz

Checkpoint quiz 3 questions
Back to course

Check your understanding of this module before moving on.

?

Checkpoint quiz 3 questions

A quick check on this module's key ideas before you move on
0/3
Question 01

What makes the credit assignment problem unique to reinforcement learning, compared to supervised learning?

B is correct. In supervised learning, every example arrives with its correct answer already labeled. In reinforcement learning, the agent only learns whether it won or lost at the end, and has to work out which of many earlier actions actually contributed to that outcome.
Question 02

Why does an agent need to explore instead of always exploiting its current best-known action?

B is correct. If the agent only ever takes the action it currently believes is best, it never gathers evidence about alternatives it hasn't tried, including ones that might actually be better.
Question 03

In the Q-learning update rule, what does the discount factor gamma control?

B is correct. A gamma close to 1 means the agent cares almost as much about distant future rewards as immediate ones. A gamma close to 0 makes the agent focus mostly on immediate reward.