CSE 471/598 Homework, Projects, and Exams


Notes: (1) Homework's and projects should be submitted before 5:00pm on the due date.
          (2) Late penalty applies with 2 (absolute) points a day. Tip: don't be late as you're busy with
               other courses too.
          (3) You can submit your homework to the instructor in class or to TA: Mr. Ryan Holmes, GWC 335.

Your suggestions are welcome. Please share your views with me (hliu@asu.edu).

HW1: (Problem Solving) Chapters 1, 2 and 3. Ex 1.2, 2.4, 2.10 (Hint: read 2.5-2.7), 3.3, 3.4
Deadline: Sept. 17 (Tues.), 2002.

HW2: (Knowledge and Reasoning) Chapters 4, 6, 7,  and 9. Ex  6.7, 6.12, 7.2 (a, b, c, d), 7.11 (a, b, c, d) , 9.4, 9.5 (a, b)
Deadline: Oct. 15 (Tues.), 2002.

HW3: (Learning) Chapter 18 and 21. Ex 18.3, 18.3 extended (create a decision tree about the move-forward for the wumpus world, and propose how you can make a learning agent for this world), 18.5, 18.7
Deadline: Oct. 31 (Thur.), 2002.

HW4: (Acting Logically) Chpaters 11, and 13. Ex 11.2 (a) and (b), 11.4 (Hint: without POP, can you do that? If you can't, why can't? You are actually discovering the anomaly!), 13.2 (The Wumpus World only)
Deadline: Nov. 21(Thur.), 2002.

HW5: (Uncertain Knowledge and Reasoning) Chapters 14, 15. Ex. 14.1 (Hint: The first principles are the definition of conditional probability), 14.6, 15.1 (a), (b), and (c)
Deadline: Dec. 10 (Tues.), 2002.


Projects

P1: (Agent and its environment)  2.5 and 3.17 (a,b,c,d)
Deadline: Oct 8, 2002. Submitted to TA: Mr. Holmes
You need to submit your code with a short report that includes the following

How to submit: A hard copy of the report should be submitted. The code can be submitted on a diskette or via an
email (if the file is very big, please talk to our TA first). You can get your submission stamped at GWC206 and leave it in our mailbox.

P2: (Learning agent) Create two learning agents: one is explained in 18.5 (in your homework 3) and the other is the decision tree based on information gain. Compare the two using the data in Fig. 18.5. You can expect that the data used in evaluation will be in the same scale and of similar data types. Both agents should output (1) the number of instances (2) the number of features, (3) deccision tree after training terminates, (4) training errors, and (5) testing errors. More data sets can be found at UC Irvine Machine Learning Repository if you are interested to find more data sets to test your learning agents.

The data format of nominal values is as follows (each data set is stored in an ASCII file as we discussed in class):

Y N E ... Y
N N W ... N
...

where the last column is about class label, every column is separated by " " and one line represents one row in Fig.18.5.

In one of the tutorials in the Powerpoint synopsis given by Ryan Holmes, there is  a string-to-list function you may find
useful http://grimpeur.tamu.edu/~colin/lp/node63.html

Deadline: Nov 19, Tues. 2002. Submitted to TA: Mr. Holmes
You need to submit your code with a short report as required in P1.

How to submit: A hard copy of the report should be submitted. The code can be submitted on a diskette. You can get your submission stamped at GWC206 and leave it in our mailboxes.
 

Exams

Two exams will cover Parts I - VI.

Exam 1 is held on Thur Oct 17, 2002.

Exam 2 is held on Thue Dec 10, 2002.

A sample program of Insert by Ryan Holmes.