Systems Theory and Automatic Control

Machine Learning, Dynamical Systems, and Control


Lecturers

Prof. Dr.-Ing. Rolf Findeisen
E-Mail: rolf.findeisen@nospam@@nospamovgu.de
Raum: G07-202

Dr.-Ing. Anton Savchenko
E-Mail: anton.savchenko@nospam@@nospamovgu.de
Raum: Room G07-205


Assistants

Maik Pfefferkorn, M.Sc.
E-Mail: maik.pfefferkorn@nospam@@nospamovgu.de
Raum: G07-209

Johanna Bethge, M.Sc.
E-Mail: johanna.bethge@nospam@@nospamovgu.de
Raum: G07-005


Time and Place

The course will take place online in an asynchronous format. Lectures and exercises will be provided via the e-learning page of this course. Additionally, we will offer regular time slots for questions and discussions regarding the lectures and exercises via zoom. Detailed information regarding the course organization as well as important announcements and information will be communicated via the course's e-learning page.

Time slot for consultations: Wednesdays, 03:00 p.m.
The consultations will take place via zoom starting on April 14. The log-in information will be provided via the e-learning page.

Tentative Course Outline:

Learning target: The lecture focuses on the fundamental concepts of machine learning with a special focus on dynamical systems and control. The listener will be enabled to understand the use and application of tools and methods from machine learning in the fields of dynamical systems and control. Special focus is put on a tailored set of methods from machine learning for the influence, understanding, analysis and control of dynamical systems. The methods and approaches are underlined by dynamical systems examples and computer exercises.

    Content:
  • Introduction to machine learning and artificial intelligence methods with a focus on dynamical systems and control
  • Regression approaches, such as Gaussian processes for the modeling and control of dynamical systems
  • Feedforward neural networks and recurrent neural networks for control and modeling of dynamical systems
  • Deep learning for modeling and control
  • Reinforcement learning and the relation to dynamic programming and optimal control

Announcements