1. True Use variable elimination to determine the probability distribution over A...

advertisement
1.
In the Bayesian Network below, the variables are A, B, C, D, E, and they are all Boolean. In the CPTs,
the notation P(a) denotes the probability that A is set to True, with similar meanings for P(b), P(c), etc.
Use variable elimination to determine the probability distribution over A given the observations that D
is False and E is True. You may eliminate variables in any order you wish, but please show your
work. (If you have neither a calculator nor time to perform the computations, then please simply write
out in full the arithmetic expressions to be evaluated.)
2.
Given the following Bayes net, build a junction tree. Perform the upward and downward passes
given the evidence that B is true and F is false. What is the distribution over A? What about over C?
A
A
T
F
P(b)
0.6
0.3
B
C
D
B
T
F
P(d)
0.7
0.1
P(a)
0.2
B
F
F
T
T
F
F
T
T
C
F
T
F
T
F
T
F
T
P(c)
0.7
0.1
F
E
A
F
F
F
F
T
T
T
T
A
T
F
P(e)
0.2
0.1
0.3
0.1
0.5
0.8
0.6
0.9
C
T
F
P(f)
0.5
0.9
3.
Given the following Bayes Net, construct the corresponding junction tree (ignore conditional
probability tables).
A
B
D
C
G
I
4.
F
E
H
J
Given the same Bayes Net in question 3, suppose we have evidence at nodes E and F, but
nowhere else. Which nodes are d-separated from node C?
Download