Module RO5501-KP04

Duration

1 Semester

Turnus of offer

each summer semester

Credit points

4

Course of studies, specific fields and terms:

• Master MES 2020, optional subject, computer science / electrical engineering
• Master CLS 2016, optional subject, computer science
• Master MES 2014, optional subject, computer science / electrical engineering

Classes and lectures:

• Graphical Models in Systems and Control (lecture, 2 SWS)
• Graphical Models in Systems and Control (exercise, 1 SWS)

Workload:

• 60 hours in-classroom work
• 30 hours private studies and exercises
• 30 hours in-classroom exercises

Contents of teaching:

• Introduction to Probability Theory, Discretely and Continuously Distributed Random Variables
• Fundamentals on Probabilistic Graphical Models
• Forney-Style Factor Graphs as a Probabilistic Graphical Model
• Message Passing via Sum- and Max-Produkt Algorithms
• Gaussian Message Passing
• State Estimation (Kalman Filtering and Smoothing including Nonlinear Extensions)
• Parameter Estimation via Expectation Maximization
• Expectation Propagation
• Control on Factor Graphs

Qualification-goals/Competencies:

• Students develop and extend their fundamental knowledge on probability theory and the transformation of discretely as well as continuously distributed random variables.
• Students can understand simple linear algorithms, such as the Kalman filter, with the help of graphical probabilistic models.
• Students can combine elements of probabilistic algorithms to novel ones with the help of graphical probabilistic models.
• Students can understand, extend and apply advanced algorithms in signal processing, parameter and state estimation as well as control to relevant problems with the help of graphical probabilistic models.

Grading through:

• written exam, oral exam and/or presentation as announced by the examiner

Responsible for this module:

• Prof. Dr. Philipp Rostalski

Teacher:

• Institute for Electrical Engineering in Medicine
• Prof. Dr.-Ing. Christian Herzog
• Prof. Dr. Philipp Rostalski

Literature:

• Loeliger, Hans-Andrea; Dauwels, Justin; Hu, Junli; Korl, Sascha; Ping, Li; Kschischang, Frank R. : The Factor Graph Approach to Model-Based Signal Processing Proc. IEEE, Vol. 95, No. 6, 2007 /li>
• Loeliger, Hans-Andrea : An Introduction to factor graphs IEEE Signal Process. Mag., Vol. 21, No. 1, 2004 /li>
• Hoffmann, Christian; Rostalski, Philipp : Current Publications from Research at the IME
• Miscellaneous : Current Publications from Research

Language:

• offered only in English

Notes:

Last Updated:

28.09.2021