CSCE 582 {=STAT 582} Bayesian Networks and Decision Graphs
See log for assignment and due dates.
HW1 1.1, 1.2, 1.3, 1.4. 20 pts--5 points per part.
HW2 1.5, 1.6, 1.7. 15 points--5 pts. per part.
HW3 1.8, 1.9, 1,10. 15 points, 5 pts per part.
1.8: (i) take off at most 2 pts; (ii):
take off at most two points for errors.
1.9: 2 points taken off per wrong
(or missing) subpart.
HW4 (Hugin=PR1) Exercise 1.11, 1.12, 1.13, 1.14, 1.15.
20 points in all. (1.11) 2 point
if the student reports
installation of Hugin (any version) on his or her PC; or, if the student
announces his or her intention of using the version in the lab; or, if the
student has done the other exercises, because this indicates that the student
can use Hugin.
(1.12) 3 points for a net file.
One point off with the wrong DAG.
Commendation for a screen dump.
Additional commendation
for two screen dumps or a screen dump showing two nets.
Additional commendation for screen dumps
indicating that evidence was entered.
(1.13, 1.14: 5 points each)
Check that three nets are
submitted. Check the structure. If wrong, take out up to three points per
exercise.
HW5 (Hugin=PR2) Exercises 2.1, 2.2, 2.3, 2.4, 2.5.
Ex. 2.1: 5 points. Check structure. If wrong, take off 3 points.
Ex.2.2: 3 points. Check structure. If wrong, take off two points.
Ex. 2.3: 3 points. If wrong answer, take off two points.
Ex. 2.4: 4 points. If wrong answer, take off two or three points.
Ex. 2.5: 5 points. Two points off if no description given. Up to two points
off if wrong DAG.
HW6. Exercise 1.15: 5 points. Incorrect structure, up to three points
off.
HW7.
Ex. 2.6, 6 points. 3 for part (i), 3 for part (ii). Check especially that the
numbers computed in part(i) are used in part (ii).
Ex. 2.7, 1 point.
Ex.2,8: (i) 2 points
The probability of infected milk is the same for each one of the
days, with the probabilities in the book. In particular, P(Inf_{i+1}=yes |
Inf_i = no) = 0.0002. Some students incorrectly used 0.0005 instead,
and they noticed instability: P(Inf1) = 0.07, and P(Inf7) = 0.13.
Take 1 point
off for this.
(ii) 3 points. Full credit for "logical" argument; full credit for algebraic
proof.
(iii) 3 points. Full credit for "logical" argument; full credit for algebraic
proof.
HW8.
Ex. 2.9, 3 points.
HW9. 10 points.
Ex. 2.18. The number of parameters to be assessed is the same. There should
be two figures, one with the network described in a way similar to figure 2.28
and one with the network similar to figure 2.29. 4 points per model, 2 points
for the conclusion that the number of parameters to be assessed is the same.
PR3.
Ex. 2.10.
(i) 10 points
(ii) 5 points
(iii) 5 points
PR4.
Ex. 2.22.
20 points in all. 4 points for each of the five parts. Go light on numerical
errors!
PR5
Total: 20 points.
Program turned in: 2 points.
Program compiles: 2 more points.
Use net file and evidence file provided by student. Runs correctly: 6 more
points.
Use asia.net and evidence as in presentation (marginal for TB, evidence A=yes
and D=yes). Runs correctly: 6 points. [0.08775096498292188,
0.9122490350170781]
Use asia.net and evidence (smoker=yes, positive X-ray = yes) and compute lung
cancer. Should obtain 64.599143, 35.400857. Runs correctly: 3 points.
Use studfarm and evidence John Sick and compute Gwenn [62.5, 37.5]: 1 point.
PR6 (extra credit)
Extra credit: 20 points
Program turned in: 2 points
Program compiles: 2 points
Use net file and evidence file provided by student. Runs correctly: 10 more
points.
If student did not provide asia with evidence in presentation as their default,
use asia.net and evidence as in presentation (marginal for TB, evidence A=yes
and D=yes). Runs correctly: 6 points. [0.08775096498292188,
0.9122490350170781] (approx.)
Otherwise, use five-node network in book, with evidence A=y, B=n, compute P(C):
6 points. [.7,.3] (approx.)