Machine Learning for Media Technology

2018/2019

Content, progress and pedagogy of the module

Objectives:

When designing and developing interactive media systems and technology, one is often faced with looking for interesting patterns and trends.This course presents theoretical concepts and practical tools for analyzing data for multimedia applications and solving machine learning problems, such as classification, in media techology. Many of these methods are used in, e.g., automatic speech recognition, face detection, web page ranking, autonomous driving, etc. The course includes the following topics: multivariate probability density functions, Bayesian classification, estimation, and detection, parametric (e.g., Gaussian density-based) and non-parametric classifiers (e.g. k-nn, parzen, convolutional neural networks), regression, data fitting, evaluation of classifiers and estimators, unsupervised  and supervised learning (e.g., reinforcement learning), feature selection and reduction.

Learning objectives

Knowledge

Students who complete the module will obtain the following qualifications:

  • Understand multivariate statistics and describe how to model multivariate data, e.g., using probabilistic and parametric descriptions

  • Understand the principles of Bayesian classification

  • Understand supervised (classification, regression) and unsupervised learning methods, (e.g., k-means clustering, principal component analysis)

  • Understand features, feature selection, and dimensionality reduction

Skills

Students who complete the module will obtain the following qualifications:

  • Choose, implement and apply pattern recognition tools to solve classification problems, e.g., footstep detection from accelerometers, recognition of single spoken digits

  • Apply knowledge to compare classification methods in terms of performance and complexity

  • Apply theory of multivariate statistics and analyze multimedia data, e.g., speech and music, images of faces, etc.

Competences

Students who complete the module will obtain the following qualifications:

  • Analyze machine learning to a problem in media technology, and reflect on a variety of possibilities to recommend a solution
  • Apply machine learning methods to this problem
  • Evaluate, discuss and generalize the results and reflect on their implications regarding the problem and the data

Type of instruction

Refer to the overview of instruction types listed in the start of chapter 3. The types of instruction for this course are decided in accordance with the current Joint Programme Regulations and directions are decided and given by the Study Board for Media Technology.



Exam

Exams

Name of examMachine Learning for Media Technology
Type of exam
Written or oral exam
In accordance with the current Joint Programme Regulations and directions on examination from the Study Board for Media Technology:

Oral or written examination with internal censor.

The assessment is performed in accordance with the 7-point scale.
ECTS5
Permitted aids
With certain aids, see list below
See semester description
Assessment7-point grading scale
Type of gradingInternal examination
Criteria of assessmentThe criteria for the evaluation are specified in the Joint Programme Regulations.

Facts about the module

Danish titleMachine learning i medieteknologi
Module codeMSNMEDM1175
Module typeCourse
Duration1 semester
SemesterAutumn
ECTS5
Location of the lectureCampus Aalborg, Campus Copenhagen, Campus Esbjerg
Responsible for the module

Organisation

Study BoardStudy Board of Media Technology
FacultyTechnical Faculty of IT and Design
SchoolSchool of Information and Communication Technology