CSC2535 Home
 
Lectures, Readings, & Due Dates
Background Readings
How to prepare project reports
Suggested Projects
Assignments
Tests
Computing
 
Geoffrey Hinton's Homepage
www.cs.toronto.edu/~hinton
 

CSC2535 -- Spring 2011
Advanced Machine Learning

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

News:

    The final test has been graded. If you send email to csc2535prof[at sign]cs.toronto.edu I will tell you your score out of 30.
    The score distribution was:
    15.5, 19.5, 20, 20.5, 22.5, 23, 23.5, 25, 25.5, 26, 27, 28.5, 28.5, 28.5, 29, 29, 30, 30

Lectures: Wednesdays 1.00-3.00pm (Room BA B024)
First lecture January 12; No lecture on February 23
Final test April 6

Instructor: Geoffrey Hinton; email csc2535prof[at sign]cs.toronto.edu
Office Hours: Wednesdays 5.00-6.00pm (Pratt 290G)

Tutor for assignment 1:
Tijmen Tieleman; email csc2535ta1[at sign]cs.toronto.edu;
Office Hours in Pratt 290C (times now confirmed by Tijmen)
5.00-6.00pm Friday Jan 28
5.00-6.00pm Friday Feb 4
4.00-5.00pm Monday Feb 7

Tutor for assignment 2:
Ilya Sutskever; email csc2535ta2[at sign]cs.toronto.edu;
Office Hours in Pratt 290C (times now confirmed by Ilya)
5.00-6.00pm Friday Feb 18
5.00-6.00pm Friday Feb 25
4.00-5.00pm Monday Feb 28

Please do NOT send email about the class directly to the personal accounts of the instructor or TA.
They will not answer class email except to csc2535prof or csc2535ta1 or csc2535ta2

Prerequisites: CSC2515: Introduction to Machine Learning (or equivalent -- ask instructor).
Load: 22 hours of lectures; two assignments; one project, one test.

Required Readings: Some of the lectures will be based on "Pattern Recognition and Machine Learning" by Chris Bishop which you should have if you have taken CSC2515. Other lectures will have a few required and optional papers or chapters per lecture. (see Lectures, Readings, & Due Dates). These readings will be available on the web.

Marking Scheme
Two assignments worth 20% each
Final test worth 20%
Project worth 40%

Computing
Many of the algorithms will be implemented in Matlab, and the projects will be done in Matlab, but prior knowledge of Matlab is not required.

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.