Faculty of Science Home Page University of Alberta Home Page

CMPUT 466/551 Schedule

See Syllabus

Please fill out this questionnaire.

Assignment Description

This is still being developed...!


65% assignments; 35% project
Di for "Dry lab only" (no coding)
Cj for "Dry lab + Coding"
Pk for "Project related"

              Assignment % (Ugrad) % (Grad)
      Due date
Bayesian Decision Theory, Linear Classifiers, Evaluating Hypotheses
B: 1,2,3,4
18 [70] 18 [90] 31/Jan 5/Feb/08
AF: 2,3,4,5
SN: 1,6, (7),8
P1 1 page Project Proposal due     5 Feb 08
Specific ML algorithms [NN, Decision Tree, SVM], Conjugate Gradient
B: 5, (6), 7, 8, 14.4
18 [65] 18 [95] 26 Feb 4/Mar/08
CW (all)
P2 "Lay of the Land" presentations     11, 13/ Mar
Belief Nets; Kernels, Gaussian Process; Lagrange Multiplier, Comparing learners + Feature Selection (Weka); Mixture Models (EM)
B: 6,8,9, E
18 [70] 18 [90] 25 27 Mar 08
SN (all)
Ensemble methods; Unsupervised learning (PCA); Computational Learning Theory; Reinforcement learning
[B: 12,14]; Reading (Sutton/Barto)
11 [65] 11 [65] Friday 11 Apr 08
CW (all)
P3 "Final" presentations     8, 10/Apr
P4 Project write-up     Thursday 24 Apr 08

Project (both Grad/Undergrad) (55%)


Scores (Gradebook): Grads UnderGrads   (Login = CNS account; password = student id)

General comments on the assignments:

Please read these carefully. Not following these overall guidelines may result in point penalty for the assignment.

  • Many assignments will includes both written questions and coding questions. Here, your submission should contain two parts: (1) A hard-copy part with your answers to the written questions and a high-level description of each of the coding questions. (2) The code itself, which must be submitted using ASTEP, before the time specified in the assignment. For more information about ASTEP, see here.
  • For the written questions:
    • Make sure your name is clearly printed on your hardcopy submission. For the sake of your privacy, do not include your student ID.
    • if you hand-write, instead of type, the written part of your assignment, you must make sure that your writing is legible.
    • you should be clear and concise ('more' does not imply 'better') and explain your answers.
    • prominently display the question number and letter.
  • For the coding questions:
    • It is up to you to convince the grader that your code works; not vice versa. This means you should include documentation, describing what the code does, and how it should work. It should also include traces that illustrate this behaviour.
    • You should follow standard 'good' coding practices: meaningful names, proper indentation, appropriate comments, etc.
    • Use the exact command-line arguments (if any), directory names and file names as described in the assignment specification so that we can run scripts to compile and test your programs automatically.
    • In general, the question will specify the language for your code (which will typically be C or C++).
    • Each coding question should be done in its own subdirectory. Typing 'make' in the subdirectory should compile the program. The name of the executable should be the same as the name of the subdirectory.
  • If you have questions about Academic Integrity, Collaboration, ... or any other policies in general, please check Policies.


In general, the labs are basically a designated time to see the TA's. This why the TAs are only there just before HWs are due. (We may have some additional labs to introduce other material. But if so, we will announce these auxiliary labs well in advance.)