Course plan

Course Objective

Along with developing skills and good understanding of the subject, the purpose of the PhD level course will be served if the students are able to come up with new research problems and solutions. The main objective is new research problem formulation, solution and eventual publication. To achieve the goal, all the students are expected to collaborate.

Course textbook

Title: PATTERN RECOGNITION AND MACHINE LEARNING. Author: CHRISTOPHER M. BISHOP. Springer Publication.

Course content

We will follow the following topics (chapters from the book):

• Introduction to the subject (Chapter 1)
• Necessary probability theory (Chapter 2)
• Linear models for regression (Chapter 3)
• Linear models for classification (Chapter 4)
• Kernel methods (Chapter 6)
• Sparse kernel machines (Chapter 7)
• Graphical models (Chapter 8)
• Mixture models and EM (expectation-maximization) (Chapter 9)
• Approximate inference (Chapter 10)

General information

Course credit: 8 points
Instructor: Saikat Chatterjee
Instructor's office:  Osquldas Väg 10, Plan 4 (floor 4).
Instructor's email: sach@kth.se
Meeting times: For discussion, mail to Instructor and fix a time
Work load: 3 hours per lecture and assignment.
Prerequisites: Good understanding of probability theory. No measure theory knowledge required.
Teaching and learning methodology: The lectures will be based on blackboard and slides. The lectures will be more discussion oriented. Students have to present papers and execute projects.

Evaluation

The evaluation criteria are based on the solving of assignments, presentation of research papers and project assignments.

Student feedback: The student feedback will be collected after the course is over and posted here. Student Feedback

Schedule and lecture notes

Class room: Teknikringen 14, room 304.
Time: 14:00 - 17:00 (Three hours per lecture)
A specific request: Please keep the instructor's mobile no: 0738913581. As the class room may be behind a locked door, In that case, call the mobile number of the instructor.

Paper presentation

For paper presentation, each group consists of two persons. Each group has to read two good papers and make a technical note followed by presentation. According to interest, students may choose quality papers. Please inform the paper title to the instructor.

Objective of technical note: For writing good technical notes, the group members should do intense discussion and exchange of ideas, revolving around the chosen papers. By this way, students are expected to learn from each other.

Tasks for technical note writing: (1) Read the paper with most attention. (2) Understand the central issues - what is the research and why and how? (3) You need to rework all the theory. (4) Then write a good summary of the research work (within 2-3 pages). (5) Present the work so that audience understands the summary. (7) Provide a thought process such that the papers can be useful for your own research.

Presentation date: 2013-08-19. If we do not finish within that day, then the next day has to be used (that is 2013-08-20). I prefer that the presentations should be finished by these dates. However, if somebody wants more time, that can be arranged in the third week of January, 2014. You may present on 2014-01-13.

Presentation time limit: Approximately one hour. Ultimately, it depends on the audience interactions.

Projects and Assignments

Objective: For projects, we are interested in hands on learning to implement existing techniques. For example, support vector machine (SVM). We will fix the project topics on the first lecture.

Group work: The projects will be excuted by forming groups. A group consists of two persons.

RVM project assignment: Please download RVM homework. Note that the assignment submission deadline is 29th July, 2013.

Project submission and presentation date: 2013-09-30. I prefer that the all the project presentations should be finished on this date. However, if somebody wants more time, that can be arranged after the third week of January, 2014, on case by case basis.

Presentation time limit: Approximately 40 minutes. Ultimately, it depends on the audience interactions.

On new research problems

Objective: The students are encouraged to formulate new research problems and solutions. The problems may or may not be associated with their main research (thesis) topics.

Mode of operation: (1) Throughout the course lectures and learning, the students should try to identify a concrete research problem where the gained knowledge and skill can be used. (2) On the last lecture class (2013-08-29), the students will go to the board or use projector to make the class undertsand about the problem formulation: motivation, the exact setup and a possible solution approach. (3) Then the student will handover a latex code and pdf document with the details. (4) The instructor compiles all the latex codes and form the bigger set of all new research problems. (5) The instructor sends the bigger note to everybody and the students are encouraged to collaborate for solutions. By forming the bigger note, everything is in one place and everybody knows the problems of others.

Hope: We will be able to solve some new interesting problems and be awarded with publications.

Another benifit: The new research problem solutions may be used for project course FEN3204 yielding 4 ECTS credits. However, the istructor is still not fully sure that whether this is permitted by KTH rules and regulations.