Contact hours
4 SWS (2 SWS exercises is self-study time; 2 SWS tutorial + 2 SWS lecture)

Credits
5 CP

Workload
150 h

Self study
90 h

Semester
Winter Semester

Cycle
every winter semester

Duration
1 Semester

Group size
60 participants

Recommended prior knowledge
Linear algebra (vectors, matrices, eigenvectors, eigenvalues, ...),
Calculus (functions, derivatives, ...),
Probability theory (joint/marginal/contidional probabilities in multiple variabels, Bayesian theorem, …).

Requirements for participation
None.

Learning Outcome
After the successful completion of this course the students
• know fundamentals of machine learning,
• know a number of important unsupervised learning methods,
• can discuss and decide which of the methods are appropriate for a given data set,
• understand the mathematics of these methods,
• can communicate about all this in English.

Content
This course first introduces into the field of machine learning and covers some basic concepts, such as learning paradigms, training, testing and generalization, and over/underfitting. It then covers a variety of unsupervised methods from classical machine learning such as principal component analysis, independent component analysis, vector quantization, clustering, Bayesian theory, and graphical models. These are shallow methods, this course does not cover deep learning methods.

Learning Methods
Lecture + self-studied exercises + inverted classroom style discussion of lecture and exercises.

Examination forms
Written final module examination of 90 minutes.

Requirements for the awarding of credit points
Passed final module examination.

Recommended literature
Lecture notes and other material are available through the learnscape (click here to get to a clickable version):