Syllabus

CSCE 768/590D Pattern Recognition and Classification

Topics:

Bayesian Decision Theory

Parametric Techniques

Maximum Likelihood Estimation

Bayesian Estimation

Principal Component Analysis

Multiple Discriminant Analysis

Hidden Markov Models

Nonparametric Techniques

Parzens Windows

Kn-Nearest-Neighbor Estimation

Linear Discriminant Functions

Relaxation Procedures

Minimum Squared-Error Procedures

Support Vector Machines

Multilayer Neural Networks

Back Propagation Algorithm

Error Surfaces

Stochastic Methods

Stochastic Search

Boltzman Learning

Evolutionary Methods

Unsupervised Learning and Clustering

Mixture Densities and Identifiability

Unsupervised Bayesian Learning

Criterion Functions for Clustering

Hierarchical Clustering

Machine Learning

Bias and Variance

Resampling

Estimating and Comparing Classifiers

For more details, stay tuned.

Reading and Lectures Policy

Attendance at lectures is mandatory. Students will be expected to have read the material for each lecture prior to the lecture and to be able to actively participate in discussions during class. In order to motivate you in this regard, there will be pop quizes every week and possibly every class. The first 5-15 minutes of the class will be spent on a pop quiz based on the assigned reading material with. You will be allowed to use notes that you have taken, however, you may not use the text book during the quiz. Keep in mind that pop quizes constitute 20% of your grade, if you are habitually late and miss these quizes you will have forfeited two letter grade.

Assignment and Project Policy

This class will entail a good deal of effort on your part. There will be regular homework/MATLAB programming projects as well as a major project. MATLAB projects must be turned in before the beginning of lecture (9:30am) on the specified due date using the dropbox program on the specified due date. We will be covering a large amount of material, and you will have a significant work load. Consequently:

No late projects or assignments will be accepted.
No extensions will be granted.

Obviously, in the event of an unforeseen emergency or serious illness such as lycanthropy, there may be an exception made. It is your responsibility to notify me of emergencies and serious illnesses in a timely manner.

Grading Policy

Participation (quizzes/HW)      20%

Individual Project              30%

Midterm Exam                    20%

Final Exam                      30%

Grade ranges

A	90 - 100	B+	86 - 89	B	80 - 85
C+	76 - 79	C	70 - 75	D+	66 - 69
D	60-65	F	below 60

Grades will not be curved. You will receive the grade that you have earned. N.B. If you want to receive a passing grade, then you must earn it during the semester. There will not be any extra credit assignments.

Cheating Policy

... Injures, therefore, should be inflicted all at once, that their ill savor being less lasting may the less offend... Niccolo Machiavelli

Assignments and examination work are expected to be the sole effort of the student submitting the work. Students are expected to follow the Code of Student Academic Responsibility found in the Carolina Community and should expect that every instance of a suspected violation will be reported. Students found guilty of violations of the Code will be subject to academic penalties under the Code in addition to whatever disciplinary sanctions are applied.

Students are expected to do their own work. While discussion of general aspects of the material is encouraged, collaborative efforts are grounds for receiving a failing grade. Academic sanctions are as follows. For the first cheating offense a student will be receive a zero on that assignment For the second offense the student will receive an F as a grade for the course.
Note: If the cheating offense is on a programming project or exam, then the student will receive an F as a grade for the course even if this is the first offense.

CSCE 768/590D Home Page

Syllabus

Homework/projects

Powerpoint Slides