This two-day course communicates basics in function optimization and its application in machine learning.
Function optimization is the bread & butter of data processing, and machine learning is directly linked to it. This two day course is meant to give you an overview of the most important techniques and how to use them. On the first day we will start slowly with objective functions and a recapitulation of linear algebra. Then we will bring the two together to understand function optimization with Quasi-Newton Methods and others.
On the second day we will have a look at more complex objective functions like neuronal networks, Markov chains and, in general, parametrised models.
The course will consist of lectures in the morning and exercises in the afternoon. To complete the exercises you will need a basic proficiency in computer programming.
Lecturer: Wolfgang Brehm
Max. number of participants: 12
This course counts for 1 SWS/semester hour in category C (key skills) of the PIER study programme.