COSC 6342: Machine Learning, Fall 2007


General Information

 

Instructor:

Ricardo Vilalta (vilalta@cs.uh.edu)

 

Office:

PGH 573

 

Telephone:

(713) 743-3614

 

Textbook:

“Pattern Recognition and Machine Learning” by Christopher M. Bishop; 1st Edition. Springer, 2006.

 

Additional Readings:

“Pattern Classification” by Duda, Hart, and Stork; 2nd Edition, Wiley-Interscience, 2000.

“Introduction to Machine Learning” by Ethem Alpaydin; 1st Edition. MIT Press, 2004.

“Machine Learning” by Tom Mitchell; 1st Edition., McGraw-Hill, 1997.

“Computer Systems that Learn” by Kulikowski and Weiss; 1st. Edition, Morgan Kaufmann,1991.

 


Information TA (Teacher Assistant)

 

Student:

Girish Nandagudi (nitinsgirish@yahoo.co.uk)

 

Office:

PGH 374

 

Office Hours:

Tuesdays 11:00 AM – 1:30 PM

 


Course Description

 

Machine Learning is the study of how to build computer systems that learn from experience. It is a subfield of Artificial Intelligence and intersects with statistics, cognitive science, information theory, and probability theory, among others. The course will explain how to build systems that learn and adapt using real-world applications from industry and science (e.g., learning to classify astronomical objects, to predict medical diagnoses, to play chess, etc.).

 

The class will be self-contained (i.e., I will not assume any previous knowledge); a review session on probability and information theory will precede those chapters in need of background knowledge. Main topics include Linear Models for Classification, Decision Trees, Neural Networks, Kernel Methods (Support Vector Machines), Graphical Models, Mixture Models and EM, Sampling Methods, Combining Models, and Reinforcement Learning.

 

Important Notice: This is a hybrid course. Lectures will only be provided on the dates marked below. Students will have access to a video recording on each class that can be downloaded through the web.

For more information visit the course on VNET.

 


Grading

 

 

Graded Work

Weight

2 Midterm Exams

60%

Homework

20%

Project

20%

 

 

The project will be a report on some area in machine learning you find most interesting.

You can either report on some novel experiments after applying an algorithm on a database or attempt a theoretical analysis.

The report must include a short survey of related work with the corresponding list of references.

 


Calendar

 

 

Dates to remember

Events

October 5

1st Midterm Exam

November 16

2nd Midterm Exam

November 21-24

No class (Thanksgiving Holiday)

November 30

Submit Project Report

Note: There is no final exam in this course.

 

 


Schedule

 

 

Dates

Topic

August 24

Introduction*

August 31

Probability and Information Theory

September 7

Linear Models for Classification

September 14

Decision Trees

September 21

Neural Networks

September 28

Kernel Methods*

October 5

Midterm Exam 1

October 12

Graphical Models

October 19

Mixture Models and EM

October 26

Sampling Methods

November 2

Combining Models

November 9

Reinforcement Learning*

November 16

Midterm Exam 2

November 30

Final Report

 

 

 *This is a hybrid course. Lectures will only be provided on the dates marked in bold.


 

 

Files for Downloading and Additional Class Information

 

Connect to the class through VNET