CS 531/ECE 517: Pattern Recognition

Instructor:

Joe Michael Kniss
Ferris 301G
277-2967
jmk (at) cs.unm.edu
Office Hours: TR 3:15-4:15 (in ECE 218)

Info:

Official class web page:
www.cs.unm.edu/~jmk/cs531

Class Blog:
unm-cs531.blogspot.com

Resources

• The Integrator : symbolic integration!
• Wikipedia : more info on topics (in english)
• Mathworld : more info on topics (in mathish)
• Slides : from the authors of the text.
• Approximating the Normal CDF (erf) and its inverse

Text:

Duda, Hart, and Stork.
Pattern Classification, 2nd edition
Wiley-Interscience, ISBN-10: 0471056693, ISBN-13: 978-0471056690
Amazon link
Errata, PDFs (by printing): Printings 1 and 2, 1and2-Section 2.11 Revision, Printings 3 and 4, Printing 5, Printing 6, Printing 7, Printing 8

Course Objectives

This course covers the fundamentals of Pattern Recognition and Classification primarily from a statististical point of view. Students will learn a variety of strategies for feature space and classifier selection, and gain exposure to a wide range of problems. Topics include Bayesian methods, parametric and nonparametric models, Markov models, kernel methods, and surveys of modern/research approaches.

Grading

Homework: 35%
Tests: 35%
Projects: 20%
Participation: 10%

Homework

Homework is to be delivered electronically, via email. Please be sure to have the words CS 531 or ECE 517 and Homework # in the email.

Policies

1. This is a graduate level class. Students are expected to conduct themselves as adults and professionals. It is the student's responsibility to keep up, ask questions, and keep the instructor informed if something prevents them from doing so.
2. Attendance   Attendance is the responsibility of the student. I will not be taking role. However, since quizes will cover the reading assinments, regular attendance is required for an A in the class.
3. Cheating   Cheating will not be tolerated! This includes copying homeworks, plagerisim, sharing information during tests, and use of unapproved packages for homeworks. A first offence will result in a 0 for the assignment/test. Repeat offences will result in an "F" for the class and potential University action.
4. Late assignments, quizes, or exams will not be accepted without prior approval from the instructor.

Schedule