Check Your Understanding

Congratulations! At this point in the lesson, you have written your own implementations of many classical dynamic programming algorithms. This is no easy feat, and you should be proud of all of your hard work!

We encourage you to take your time with this content. Tinker more with the mini project to develop your intuition, and read Chapter 4 (especially 4.1-4.4) of the textbook to supplement your understanding.

You are strongly encouraged to take your own notes. You may find it useful to compare your notes with the next concept, which contains a summary of the main ideas from the lesson.

When you're ready, answer the question below to check your memory of the terminology.

QUIZ QUESTION::

Match each algorithm to its appropriate description.

ANSWER CHOICES:

Finds the optimal policy through successive rounds of evaluation and improvement.

Given a value function corresponding to a policy, proposes a better (or equal) policy.

Computes the value function corresponding to an arbitrary policy.

Finds the optimal policy through successive rounds of evaluation and improvement (where the evaluation step is stopped after a single sweep through the state space).

Description	Algorithm
Policy Improvement
(Iterative) Policy Evaluation
Value Iteration
Policy Iteration

SOLUTION:

Description	Algorithm
Policy Improvement	Given a value function corresponding to a policy, proposes a better (or equal) policy.
(Iterative) Policy Evaluation	Computes the value function corresponding to an arbitrary policy.
Value Iteration	Finds the optimal policy through successive rounds of evaluation and improvement (where the evaluation step is stopped after a single sweep through the state space).
Policy Iteration	Finds the optimal policy through successive rounds of evaluation and improvement.