ELEN E6880              Spring 2002
Statistical Pattern Recognition

Homework Sets 

Homework #1
Homework #2
Homework #3
Homework #4
Homework #5

Project Home Page
Teams for the Project


Lectures 

The material for the lectures include foils, lecture notes, links to additional material, and references. 

1. Probability
2. Introduction
3. Linear Discriminants 
4. Neural Networks
5. Parametric Methods
6. Kernel Methods 
7. Feature Selection 
8. Nearest Neighbor 
9. Tree Classifiers 
10. SVMs 
11. Boosting, etc. .
12. Bayesian Nets 
13. Unsup. Learning
14. Comparing

Instructors
Vittorio Castelli
vittorio@us.ibm.com
(914) 945-2396
Mark Brodie
mbrodie@us.ibm.com
(914)784-7484
Accomplices
Irina Rish
rish@us.ibm.com
(914)784-7431
Daniel Oblinger
oblinger@us.ibm.com
(914) 945-2326
Grader
Joseph Yangi Yun      yy49@columbia.edu     Office: Mudd 244A   (Phone) 212-854-3155


The class meets Wednesday, 6:50 - 9:20 PM, in Mudd 545.
Office Hours with the instructors are Wednesday, 5:50-6:50 PM, in Mudd 1319.
Office Hours with our grader are on Thursday, 10 AM-12 PM, in Mudd 244A.

Do not hesitate to contact us via e-mail, should you have questions!

Course Information (Courtesy of CVN), Including Time Table



Announcements
04/26/2002 -  New course material
04/26/02 Unsupervised Learning page updated, with relevant chapters and bibliography.
04/26/02 The slides for the Bayes Network lecture are on-line
04/22/02 Unsupervised Learning Lecture Notes Available
04/19/02 Homework set nr 5 is now available
04/12/02 A paper with material for Lecture 12 is available in the additional material

Useful Links

 

Grading Policy

There are 4 components to the grade for the class
  • Homework:  30%
  • Midterm:       25%
  • Final:             35%
  • Project:         15% for solving the first part, 10% for solving the second part
Note that the overall grade without the second part of the project is 105%, namely, the project gives student 5% bonus points.  Students desiring to do additional work to improve their grade are encouraged to try part 2 of the project.   The grading policy for this second part of the project is going to be more strict than the first part: in other words, only students who are really serious about the additional work should undertake it.
     

    Last Modified 04/08/2002