EE573

EE573 Pattern Recognition 

Fall 2017   Course Information

The aim of this course is to introduce the students to the fundamental problems  in  pattern recognition and familarize them with the most common approaches to these problems.  Topics include Bayesian Decision Theory,  Parametric Estimation, Discriminant Methods and  Unsupervised Learning. The course emphasizes both the theoretical and practical aspects of these various approaches.

Instructor  Prof. H. Isil Bozma

Teaching Assistant:  Bayram Akdeniz -email:  bayramakdeniz89 at gmail dot com

Class

Lectures: Mondays 12-14 @ Shannon, Wednesdays 13-14@  Shannon

Prerequisites :  Probability, working knowledge of Matlab & Simulink,  C or C++ or Java

Textbooks

Duda, Hart, Stork, Pattern Classification, 2nd Ed., Wiley-Interscience, 2004

Grading

Grading will be based on projects, one midterm  and a final. I will also expect you to attend the lectures and contribute to in-class discussions.The weights of each will  announced asap.

Projects Midterm Final Class Participation

Please note that only students with passing status on projects, midterm and class participation will be allowed to take the final.

Minimum expected class attendance: 75% of all the lectures.

Syllabus  Fall 2017

Please note that this is a tentative plan  and may change throughout the semester based on our progress.

  • Week  1 (18 – 22  Sept   ) 
  • Week  2 (26 – 30 Sept  ) 
    • No lectures this week
  • Week 3 ( 2- 6  Oct)
    • Bayesian Decision Theory (DHS Ch 2)  –
    • hw1
  • Week 4 ( 9 – 13  Oct)
  • Week 5 (16 – 20 Oct)
  • Week 6 (23 – 27 Oct )
  • Week 7 (30 Oct – 3 Nov )
    • Lecture: Markov Models  DHS Chapter 3.10
    • Linear Discriminant Methods (DHS Ch5)
  • Week 8 (6 – 10 Nov)
    • Linear Discriminant Methods (cont.)
    • Midterm
  • Week 9 (13 – 17 Nov)
    • Linear Discriminant Methods (DHS Ch 5 ) 
    • Supplementary material: Ho-Kashyap
    • Support vector machines  SVM
  • Week 10 (20  – 24  Nov )
    • Parametric Methods (DHS Ch 4)  –  Parzen Window
  • Week 11 ( 27 Nov – 1  Dec )
    • DHS Ch4 (Part 2) –  k-Nearest Neighbor
  • Week 12 (4 – 8 Dec)
    • Tree methods (DHS Ch 8) 
  • Week 13 (11 – 15 Dec)
    • Unsupervised decision making

Projects

You will be required to homeworks/projects for the course.  They may have some theoretical and coding parts.

  • Pls hand in the theoretical parts in written form – including the short project report if you are asked to do so.
  • Pls send your source codes + a readme file via email both to me and the course TA. BE SURE TO REFER TO ANY LITERATURE/EXTERNAL CODE you have examined or used. A separate REFERENCE section must be added to the end of all your reports – whenever applicable. Again, I assume that the projects have not been or are currently being done for other courses, etc.

In the projects, you mau use the code available in the DHS Matlab toolbox.