Time and place
Lecture: Di, Do, 12-14h, HS II, Albertstr. 23b
Tutorial: 2 hours, date to be determined
Requirements on examinations, assessments and coursework will be described in the supplements of the module handbooks to be published as part of the course cataloque by end of October.
Teaching
Teacher: Giuseppe Genovese
Language: in English
Content
The goal of the course is to provide a mathematical treatment of deep neural
networks and energy models, that are the building blocks of many modern machine learning
architectures. About neural networks we will study the basics of statistical learning theory, the
back-propagation algorithm and stochastic gradient descent, the benefits of depth. About energy models we will cover some
of the most used learning and sampling algorithms.
In the exercise classes, besides solving theoretical problems, there will be some Python programming sessions to implement the models introduced in the lectures.
Previous
knowledge
Probability Theory I \
Basic knowledge of Markov chains is useful for some part of the course.
Usability
Elective (Option Area) (2HfB21)
Compulsory Elective in Mathematics (BSc21)
Applied Mathematics (MSc14)
Mathematics (MSc14)
Concentration Module (MSc14)
Elective (MSc14)
Advanced Lecture in Stochastics (MScData24)
Elective in Data (MScData24)