389.207 Bayesian Machine Learning
This course is in all assigned curricula part of the STEOP.
This course is in at least 1 assigned curriculum part of the STEOP.

2023W, VO, 2.0h, 3.0EC


  • Semester hours: 2.0
  • Credits: 3.0
  • Type: VO Lecture
  • Format: Presence

Learning outcomes

After successful completion of the course, students are able to understand and apply fundamental theory and methodology of Bayesian machine learning. A further outcome is an improvement of English language skills.

Subject of course

* Introduction:  Motivation, applications, outline.

* Bayesian estimation:  General Bayesian estimator, MMSE estimator, MAP estimator, ML estimator, application example.

* Bayesian classification:  General Bayesian classifier, MAP classifier, ML classifier, application examples.

* Exponential family:  Definition and expressions, log-partition function, sufficient statistic, ML estimator, posterior distribution, conjugate prior, MMSE and MAP estimators, examples.

* Bayesian networks:  Definition, basic examples, conditional independence, d-separation property, Markov blanket and boundary.

* Variational Bayesian inference: Laplace approximation, evidence lower bound, mean-field approximation, CAVI algorithm, exponential family model, model with global and local parameters, application example: word-topic modeling using latent Dirichlet allocation, stochastic variational inference, expectation propagation.

* Latent variable methods:  EM algorithm, MAP-EM algorithm, variational EM algorithm, auto-encoding variational Bayes method, variational autoencoder.

Teaching methods

The prof (Hlawatsch) verbally presents the class material, discusses the material with his students, and answers the students' questions. For this, he uses a blackboard, on which he writes certain characters and draws simple figures with a piece of chalk (also using different colors if helpful). He also uses a tablecloth to erase the board every now and then. Finally, he uses an overhead projector to project more complicated figures and tables on a screen. The prof's presentation is supported by lecture notes.

Mode of examination


Additional information

This course is an optional part of the "Wahlmodul Advanced Signal Processing."

First class: Thursday, 5 October 2023, 3.15 pm in seminar room 118 (Sem. 389). The course will take place in presence mode.



Course dates

Thu15:00 - 17:3005.10.2023 - 25.01.2024Sem 389 Bayesian Machine Learning
Bayesian Machine Learning - Single appointments
Thu05.10.202315:00 - 17:30Sem 389 Bayesian Machine Learning
Thu12.10.202315:00 - 17:30Sem 389 Bayesian Machine Learning
Thu19.10.202315:00 - 17:30Sem 389 Bayesian Machine Learning
Thu09.11.202315:00 - 17:30Sem 389 Bayesian Machine Learning
Thu16.11.202315:00 - 17:30Sem 389 Bayesian Machine Learning
Thu23.11.202315:00 - 17:30Sem 389 Bayesian Machine Learning
Thu30.11.202315:00 - 17:30Sem 389 Bayesian Machine Learning
Thu07.12.202315:00 - 17:30Sem 389 Bayesian Machine Learning
Thu14.12.202315:00 - 17:30Sem 389 Bayesian Machine Learning
Thu21.12.202315:00 - 17:30Sem 389 Bayesian Machine Learning
Thu11.01.202415:00 - 17:30Sem 389 Bayesian Machine Learning
Thu18.01.202415:00 - 17:30Sem 389 Bayesian Machine Learning
Thu25.01.202415:00 - 17:30Sem 389 Bayesian Machine Learning

Examination modalities

Oral exam

Course registration

Not necessary


Study CodeObligationSemesterPrecon.Info
066 507 Telecommunications Not specified


Lecture notes can be downloaded -- see links further below.

Recommended textbook: Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer, 2006.  Free download:


Previous knowledge

Working knowledge of random variables and linear algebra