ECE 788: Pattern Recognition, Machine Learning and Information Theory


Description

This course provides a systematic introduction to pattern recognition and machine learning using information-theoretic performance criteria as guiding principles. Topics covered include linear and Kernel models for classification and regression, sample complexity and VC dimension, probabilistic graphical models and approximate inference.

Prerequisites

Good knowledge of probability and algebra.

Instructor

Dr. Osvaldo Simeone
Email: osvaldo.simeone@ njit.edu
Phone: (973) 596-5809
Office: 101 FMH Building 
Office Hour: Wednesday 4-6pm

Textbooks and Lecture Notes

See also


Requirements

There will a midterm (40%), a final exam (40%), scribe or project (20%).

Assignments use MATLAB (see here for an introduction)


Tentative schedule

Week

Plan

Chapter covered

1,2

Introduction (linear regression)

B.1, B.3

3,4

Learning and probability

B.2

5

Linear models for classification

B.4, B.6

6

Statistical learning

S.2, S.3, S.5, S.6

7

Midterm

 

9,10

Unsupervised learning

B.9  

11,12

Probabilistic graphical models

B.8

13,14

Approximate inference

B.10, B.11

15

Final

 

 


NJIT Honor Code

The NJIT Honor Code will be upheld, and any violation will be brought to the immediate attention of the Dean of Students.

 

Changes in the syllabus might be possible. Students will be informed of those changes in the class announcements and on the web.