Winter Term
Deep Learning
Course number: 212018
Type: Lecture
CP: 5
SWS: 4
Language: English
Description: Deep learning is a subfield of machine learning, which has led to breakthroughs in numerous application areas (such as object and speech recognition and machine translation) in recent years.
The aim of the lecture is to provide an insight into this field. At the beginning, the basic terms and concepts of machine learning are introduced. In the further course, various neural networks, gradient-based optimization methods and generative models are discussed.
Deep learning methods are used in areas such as IT security.
Discrete Mathematics
Course number: 212063
Type: Lecture
CP: 8
SWS: 6
Language: German
Summer Term
Advanced Topics in Deep Learning
Course number: 212120
Type: Seminar
CP: 3
Language: English
Description:
Deep learning is the field of machine learning based on artificial neural networks. In the last decades deep learning has revolutionized a lot of applied fields like computer vision, speech recognition, natural language processing, machine translation, bioinformatics, and many more.
In this Master Seminar we will read and present recent publications in the field of deep learning. Students should have some background knowledge on machine learning and deep learning (and ideally having taken a lecture on deep learning). Presentations will be organized in small blocks at the end of the semester.
Introduction to Artificial Intelligence
Course number: 211045
Type: Lecture
CP: 5
SWS: 4
Language: English
Description: This course gives an overview over representative methods in artificial intelligence: formal logic and reasoning, classical methods of AI, probabilistic reasoning, machine learning, deep neural networks, computational neuroscience, neural dynamics, perception, natural language processing, robotics.
Python Programming and Basic Machine Learning
Course number: 211434
Type: Lab Course
CP: 3
SWS: 4
Language: English
Description: The course will include 4 main parts. In the first part the basic coding skills would be deepened with possible filling the gaps in the knowledge of the students. Various practical problems will be implemented and ran. Second part will be devoted to the work with data and basic data analysis, with main emphasis on using Pandas library. Third part will be a short introduction to basics of machine learning and machine learning models that can be implemented and trained using scikit-learn library. Final part is basic deep learning using PyTorch library. The self-work will be devoted to a course project which will be implementation of a neural network for a particular task and work with it. The work in the course project has to be summarized in a report and a presentation for successful finishing the course.
Theory of Machine Learning
Course number: 211052
Type: Lecture
CP: 9
SWS: 6
Language: German
Description: Gegenstand der Vorlesung ist die Statistik- und Algorithmen-basierte Theorie des Maschinellen Lernens aus zufälligen Beispielen. Wir befassen uns mit der Bestimmung der Informations- und der Berechnungskomplexität von Lernproblemen. Im ersten Teil der Vorlesung behandeln wir die grundlegenden Begriffe und Resultate der Theorie des maschinellen Lernens. Im zweiten Teil der Vorlesung beschäftigen wir uns mit verschiedenen Ansätzen zum Design von maschinellen Lernalgorithmen (wie zum Beispiel Boosting, stochastischer Gradientenabstieg, kernbasierte Verfahren, Entscheidungsbäume, Nearest Neighbor).