Skip to main content
Till KTH:s startsida

FEP3302 Majorize-Minimization (MM) Optimization with Machine Learning Applications 7.0 credits

Introduce the theory of MM principle with an emphasis on related optimization algorithms and their applications to machine learning.

Information per course offering

Course offerings are missing for current or upcoming semesters.

Course syllabus as PDF

Please note: all information from the Course syllabus is available on this page in an accessible format.

Course syllabus FEP3302 (Spring 2022–)
Headings with content from the Course syllabus FEP3302 (Spring 2022–) are denoted with an asterisk ( )

Content and learning outcomes

Course disposition

Introduction (Lecture 1)

o MM principle

o A geometric interpretation

o Convexity for Majorization

o Examples

Key Inequalities for MM (Lecture 2 and 3)

o Applications of Jensen’s inequality

o Applications of the Cauchy-Schwarz inequality

o Applications of supporting hyperplane inequality

o Application of quadratic upper bounds

o Application of arithmetic-geometric mean inequality

Majorization and Partial Optimization (Lecture 4)

o Main principle

o Examples

Application in Engineering (Lecture 5 and 6)

o EM algorithm

o Regression

o Estimation with missing data

o Total variation denoising of images

o Factor analysis

o Matrix completion

Course contents

 Introduction (Lecture 1)

o MM principle

o A geometric interpretation

o Convexity for Majorization

o Examples

Key Inequalities for MM (Lecture 2 and 3)

o Applications of Jensen’s inequality

o Applications of the Cauchy-Schwarz inequality

o Applications of supporting hyperplane inequality

o Application of quadratic upper bounds

o Application of arithmetic-geometric mean inequality

Majorization and Partial Optimization (Lecture 4)

o Main principle

o Examples

Application in Engineering (Lecture 5 and 6)

o EM algorithm

o Regression

o Estimation with missing data

o Total variation denoising of images

o Factor analysis

o Matrix completion

Intended learning outcomes

LO1: Recognize the concept of MM Principle.

LO2: Incorporate techniques for majorization and minorization into the design of MM

optimization algorithms.

LO3: Implement numerically the MM optimization algorithms in various application.

Literature and preparations

Specific prerequisites

 Multi variable analysis, probability theory

Recommended prerequisites

 Multi variable calculus, probability theory

Literature

You can find information about course literature either in the course memo for the course offering or in the course room in Canvas.

Examination and completion

Grading scale

P, F

Examination

  • EXA1 - Examination, 7.0 credits, grading scale: P, F

Based on recommendation from KTH’s coordinator for disabilities, the examiner will decide how to adapt an examination for students with documented disability.

The examiner may apply another examination format when re-examining individual students.

If the course is discontinued, students may request to be examined during the following two academic years.

A take-home exam (4-5 problems) or/and group presentations of a simple implementation

Examiner

Ethical approach

  • All members of a group are responsible for the group's work.
  • In any assessment, every student shall honestly disclose any help received and sources used.
  • In an oral assessment, every student shall be able to present and answer questions about the entire assignment and solution.

Further information

Course room in Canvas

Registered students find further information about the implementation of the course in the course room in Canvas. A link to the course room can be found under the tab Studies in the Personal menu at the start of the course.

Offered by

Education cycle

Third cycle

Postgraduate course

Postgraduate courses at EECS/Network and Systems Engineering