CSC321 Winter 2014 - Course information
Introduction to Neural Networks and Machine Learning

This course is taught using the "inverted classroom" model. This means that instead of being introduced to the material in a largely one-way lecture in a hall, you'll watch the lecture as a video at home before class, and then in class, we can have a much more dynamic discussion about it.

Each lecture has been recorded by Geoffrey Hinton as a set of about 3 short videos. If you are registered for the class, you will be able to view these videos on the Coursera website. Details of how to do this will be given in the first lecture period. Each lecture period will consist of a class discussion of the content of the videos for that lecture, so you must watch the videos before the corresponding lecture period. If you don't, most of the discussion will probably be lost on you and class will be wasted time.

Lectures: Tuesdays and Thursdays, 1:10 - 2:00 pm, in University College, room 179.
The first lecture is on January 7; the last lecture is on April 3. There are no lectures on February 18 and 20 (reading week).
See the calendar page for a list of all the lectures (subject to minor change).
Note that the lecture periods will not be lectures in the traditional sense, because of the "inverted classroom" model that we're using.

Tutorials: Thursdays 12:10 - 1:00 pm (also in UC 179).
Some weeks there is no tutorial. See the calendar page for details of the tutorials.

Teaching team:
  • Instructor: Tijmen ("Tim") Tieleman (email csc321prof[at sign]cs.toronto.edu).
    Office hours: Fridays 2-3pm, in BA 3201 (starting January 10; last time is April 4).
    Of course you can also talk to me before or after class, or send me an email.
  • Voice of Wisdom, sometimes in class and always in the videos: Geoffrey Hinton.
  • Teaching Assistants:
    • Nitish Srivastava
    • Chris Maddison
    • Jian Yao
    • Hannes Bretschneider

Do NOT send us email about the class directly to our personal accounts. We will not answer.

Prerequisites: If you want a waiver, you MUST contact Tijmen in the first week.
The prerequisites that will be enforced are:
  • (MAT135H1, MAT136H1)/MAT135Y1/MAT137Y1/MAT157Y1
  • MAT223H1/MAT240H1
  • STA247H1/STA255H1/STA257H1
  • CGPA 3.0, or enrollment in a CSC subject POSt


Load: 24 hours of lectures; 11 hours of Tutorials.

Required Readings: There is no required textbook for the class.
Most of the course material is presented in the lecture videos. A few small readings may be assigned if the need arises. These required readings will all be available on the web, for free.

Marking Scheme
  • Midterm test: 20%.
  • Final exam: 35%.
  • Four assignments worth 10% each.
  • The quizzes on the Coursera website: 5%.
By the time you get to an advanced course like csc 321 you've heard this lots of times, so I'll keep it brief: avoid academic offences (a.k.a. cheating). All graded work in this course is individual work.

Computing
The assignments will all be done in Matlab, but prior knowledge of Matlab is not required. Basic Matlab will be taught in a tutorial.

Online forum
We'll use the forum that's part of the Coursera website for this course.

Coursera
Lecture material, quizzes, and the forum for this course are hosted on the Coursera website. See the page about Coursera.

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.

Map of Campus Buildings

[ Home | Course information | Lecture notes | Quizzes | Optional Readings | Computing | Assignments | Tests | Coursera ]

CSC321 - Introduction to Neural Networks and Machine Learning