Question 1 (4 points)

Which of these are examples of reasons to prefer reasoning in terms of features rather than in terms of states when addressing a problem?

Select all that apply.

Question 1 options:

1)	The number of features necessary to reason about a system is often many fewer than the number of states in such a system.

2)	Features are independent of each other.

3)	It is often more natural to reason about a system in terms of its features than in terms of its states.

4)	States can be defined in terms of features, but features cannot be defined in terms of states.

Question 2 (2 points)

Which of these are elements of a Constraint Satisfaction Problem? Select all that apply.

Question 2 options:

1)	A set of variables

2)	A set of features

3)	A set of states

4)	A set of constraints

Question 3 (1 point)

True or false: A Generate-and-Test algorithm can solve any finite Constraint Satisfaction Problem. (That is, if there is a satisfying assignment, G&T will find it, and if there is no satisfying assignment, G&T will so state).

Question 3 options:

	1) True
	2) False

Question 4 (4 points)

Assume that you post a Constraint Satisfaction problem (CSP) for which no solution (a.k.a., "no model") exists. Which of the following methods can be used to determine with certainty that no model exists. Select all that apply.

Question 4 options:

1)	Generate-and-test

2)	Search with backtracking

3)	Arc consistency (without domain splitting)

4)	Variable elimination

Question 5 (4 points)

Assume that you post a Constraint Satisfaction problem (CSP) for which multiple solutions (a.k.a., "multiple models") exist. Which of the following methods can be used to find all the models that exist. Select all that apply.

Question 5 options:

1)	Generate-and-test

2)	Search with backtracking

3)	Arc consistency (without domain splitting)

4)	Variable elimination