CS486/686 - Assignments

There will be four assignments, each worth 10% of the final mark (7% for CS686). Each assignment will have a theoretical part and a programming part.  Assignments are done individually (i.e., no team).  You are free to program in the language of your choice, however Python and Matlab are recommended since they provide a convenient high-level programming environment for matrix operations.  If you decide to program in Matlab, the IST group maintains a nice set of online references for Matlab including a tutorial.

The approximate out and due dates are:

• A1: out May 12, due May 27 (11:59 pm)
  
• A2: out May 28, due June 12 June 15(11:59 pm)
  
• A3: out June 23, due July 8 (11:59 pm)
• A4: out July 9, due July 24 (11:59 pm)

• Click here for the assignment
• Click here for the test problems.  The problems are taken from www.websudoku.com and are labeled "easy", "medium", "hard" and "evil" to reflect the category that each problem is taken from.  Note that it is not clear how those labels are assigned and therefore they may not reflect the level of difficulty encountered by search algorithms.
• Submit a pdf file or zip file that contains your code and answers to each question in the dropbox for assignment 1 on LEARN.  If you are not familiar with LEARN follow the instructions posted here.
  
• Milad Khaki (mkhaki [at] uwaterloo [dot] ca) is the TA in charge of Assignment 1.  He will hold special office hours on Friday May 22, 2-4pm in the AI lab (DC2306C).

Assignment 2: due June 12 June 15 (11:59 pm)

• Click here for the assignment.
• Hadi Hosseini (h5hossei [at] uwaterloo [dot] ca) and Shengying Pan (s5pan [at] uwaterloo [dot] ca) are the TAs in charge of Assignment 2. They will hold special office hours on Tuesday June 9, 10am-noon in the AI lab (DC2306C).

Assignment 3: due July 8 (11:59 pm)

• Click here for the assignment
• Train and test your algorithms with a subset of the 20 newsgroup dataset.  More precisely, you will use the documents posted on the alt.atheism and comp.graphics newsgroup.  To save you the trouble of writing a parser for arbitrary text, I converted the relevant documents to a simple encoding (files below).  Each line of the files trainData.txt and testData.txt are formatted "docId wordId" which indicates that word wordId is present in document docId.  The files trainLabel.txt and testLabel.txt indicate the label/category (1=alt.atheism or 2=comp.graphics) for each document (docId = line#).  The file words.txt indicates which word corresponds to each wordId (denoted by the line#).  If you are using Matlab, the file loadScript.mprovides a simple script to load the files into appropriate matrices.  At the Matlab prompt, just type "loadScript" to execute the script.  Feel free to use any other language and to build your own parser if you prefer.
• trainData.txt
• testData.txt
• trainLabel.txt
• testLabel.txt
• words.txt
• loadScript.m
• Daniel Patrick Recoskie (dprecosk [at] uwaterloo [dot] ca) and Milad Khaki (mkhaki [at] uwaterloo [dot] ca) are the TAs in charge of Assignment 3.  They will hold special office hours on Friday July 3, 11:30am-1:30pm in the AI lab (DC2306C).

Assignment 4: due July 24 (11:59 pm)

• Assignment handout
• Datasets: trainData.csv, trainLabels.csv, testData.csv, testLabels.csv
• This data is a modified version of the Optical Recognition of Handwritten Digits Dataset from the UCI repository.  It contains preprocessed black and white images of the digits 5 and 6.  Each attribute indicates how many pixels are black in a patch of 4 x 4 pixels.
  
• Format: there is one row per image and one column per attribute.  The class labels are 5 and 6.
• In Matlab, you can easily load a file by typing "load filename.csv"
  
• Wei-Shou Wilson Hsu (wwhsu [at] uwaterloo [dot] ca) and Shengying Pan (s5pan [at] uwaterloo [dot] ca) are the TAs in charge of Assignment 4. They will hold special office hours on Wednesday July 22, 3-5pm in the AI lab (DC2306C).