| |
CSC2515 Fall 2007 - Lectures
Lecture Schedule:
This course is being changed a lot from last year.
The lecture titles below are all provisional at present and the
lectures do not yet exist.
The FINAL VERSION of each
lecture and the FINAL VERSION of the readings for that lecture
will be posted on or before
the day of the lecture.
- September 11
Lecture 1: (11.00-1.00) Overview of Machine Learning
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Reading: Chapter 1, pp 1-48.
Tutorial 1: (1.00-2.00) The Gaussian Distribution
Reading: Chapter 2, pp 78-94
- September 18
Lecture 2: Linear Regression
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Reading: Chapter 3, pp 137-173
Tutorial 2: (1.00-2.00) Distributions for binary and multinomial variables; The exponential family.
Reading: Chapter 2, pp 67-78; 113-120.
- September 25
Lecture 3: Linear Classification
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Reading: Chapter 4: pp 179-210
Tutorial 3: (1.00-2.00) Worked examples using the material covered so far.
- October 2
Lecture 4: Neural Networks trained by Backpropagation
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Reading: Chapter 5: 225-249, 256-269
- October 9 :
Lecture 5: Clustering and Mixture Models
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Tutorial: (1.00-2.00) Help with Matlab for novice users. Note: The TA
will not tell you what program to write for the assignment!
Reading: Chapter 2: 120-127; Chapter 9: Pages 423-455; Chapter 5:269-272
- October 16 Assignment 1 due at start of
lecture
Lecture 6: Optimization methods for Machine Learning
Tutorial: 1.00-2.00 Introduction to assignment 2.
(notes
as .ppt )
(notes
for all browsers))
(pdf
notes, 6 to a page))
Optional Reading:
Conjugate
gradient paper
- October 23 One page project proposal due (see
projects page)
Lecture 7: Decision Trees and Mixtures of Experts
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Readings: Chapter 5: 272-277
Mixture of
Experts paper
Hierarchical Mixture of
Experts paper
Lecture 7b: Kernel density estimator and nearest neighbors
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Reading: Chapter 2: 120-127
- October 30 Assignment 2 due at start of lecture
Lecture 8: Deep Belief Nets
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Optional Reading List
- November 6
Lecture 9: Continuous Latent Variable Models
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
- November 13
Lecture 10: Time-series Models
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Reading:
- November 20 Assignment 3 due at start of lecture:
Lecture 11 Combining Models
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Reading:
- November 27
Lecture 12 Kernel Methods
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Reading:
- December 4 Assignment 4 due at start of lecture
Lecture 13 Bounds on Generalization
(notes as .ppt )
(notes for all browsers))
(notes as .ps, 4 per page))
Reading:
- Tuesday December 18 Projects due before midnight.
Email .pdf or .ps file to csc2515prof@cs.toronto.edu
|