Due Tuesday, October 10 at 11:59 PM via Canvas. PDF format required.

1. (1 points) Exercise 3.1

2. (2 points) Exercise 3.5

3. (2 points) Exercise 3.6

4. (1 point) Exercise 3.12a (Clarification: the arc being reversed is between Bulglary and Alarm.)

5. (2 points) Exercise 4.3 (Clarifications: H_n contains n nodes. You want to show that the size of the parameter space of the Bayesian Network is exponential in n.)

6. (1 point) Exercise 4.10 (only the strong union part)

7. (1 point) For the network in Figure 3.15, consider a "smaller" version that includes only the six random variables other than Alarm. In order to ensure the six-variable model captures all the dependencies between those six variables that are present in the original model, how many edges would you need to add? Assuming all random variables are binary, how many parameters are there in the original seven-variable model? In the six-variable model?