CSC2515 -- Fall 2008
Introduction to Machine Learning

Students should check these pages at least once a week for important information.

News:
CDF labs will be closed for the holidays from 9.00am on dec 19 to monday Jan 5, but the system will be available for remote logins. So dont leave your computations until the last 3 hours if you are
using CDF.
Information about how to get multiclass SVMlite code for the DBN project is now on the projects page.



Lectures and Tutorials: Wednesdays 1.00-3.00, 3.00-4.00 in room BA2185
The three hour period will typically be used for a two hour lecture followed by a one hour tutorial.
First lecture Wednesday September 10

Instructor: Geoffrey Hinton; email csc2515prof[at sign]cs.toronto.edu
Office Hours: Geoffrey Hinton: Wednesdays 5.00-6.00pm (Pratt 290G) until Nov 26

Tutors: Andriy Mnih and Russ Salakhutdinov; email csc2515ta[at sign]cs.toronto.edu
Office Hours: Thursdays 3.00-4.00pm (Pratt 275, Pratt 265D)

Please do NOT send the instructor or tutors email about the class directly to their personal accounts. It will be deleted. We will only respond to class email if it is sent to csc2515prof or csc2515ta[at sign]cs.toronto.edu

Prerequisites: some knowledge of calculus, linear algebra, and statistics
Load: About 26 hours of lectures; three assignments; one project; (plus one final test).

Required Readings: The textbook for the course is "Pattern Recognition and Machine Learning" by Chris Bishop. This will be available in the UofT bookstore. Most of the required readings will be parts of the textbook but there will be some additional readings and we will not cover all of the chapters in the textbook. (see Lectures, Readings, & Due Dates) for details of the required readings.

Marking Scheme
Three assignments worth 15% each
Project worth 30%
Final test worth 25%

Computing
Parts of the assignments will be done in Matlab, but prior knowledge of Matlab is not essential.

Auditing
If you are not registered in the class, it is possible for you to audit it (sit in on the lectures), but only if you get the instructor's permission and follow some rules. See the audit page for more info.