LFU:online

Prerequisites for registration: none

Learning Outcome: Students of this module understand the content of the lectures and they can reproduce and apply it. They have acquired the ability to produce similar content by themselves. They can identify optimisation problems in applications and transfer them into a mathematical formula and implement the corresponding algorithms and solution procedures appropriately. Moreover, they should have gained a fundamental understanding of the methods of mathematical optimisation.

Prerequisites for registration: none

Learning Outcome: Students who have completed that module have acquired particular knowledge of one or more branches of higher mathematics. They are able to develop innovative solutions for current problems of those branches of mathematics as well as to judge different approaches critically. As a result they have developed learning strategies that enable them to acquire further mathematical matters autonomously.

Prerequisites for registration: none

Learning Outcome: Students who have completed that module have acquired a deepened knowledge of a branch of higher mathematics by autonomous studies. Moreover they are familiar with relevant mathematical literature and can judge its mathematical content. They are able to examine problems of higher mathematics in a creative and methodically correct manner and to present the result of those examinations in written form and orally as to be understood well by experts. The contents of the seminars are oriented on current research topics.

Prerequisites for registration: none

Learning Outcome: Having successfully completed this module, students are able to actively participate in the discussion of the current state of knowledge in the area of the dissertation topic and can critically reflect on and discuss issues with experts of the chosen partial discipline. On this basis, they are able to make their own contributions to research.Likewise the students have interface knowledge at a high professional level which is needed for the implementation of the dissertation.

Prerequisites for registration: absolvierte Lehrveranstaltungen im Umfang von mindestens 15 ECTS-AP aus Pflichtmodul 1

Learning Outcome: Studierende erwerben tiefgehende Kenntnisse über Supervised Learning mittels Verteilungsregression, über Unsupervised Learning für multivariate Daten sowie Supervised Learning mittels algorithmischer Modelle. Sie besitzen die Fähigkeit, für eine konkrete Aufgabe eigenständig ein wahrscheinlichkeitstheoretisches Modell auszuwählen und anzupassen, insbesondere fällt darunter die Wahl geeigneter Antwortverteilung, Regressoren und Algorithmen zur Parameterschätzung. Sie sind in der Lage, dimensionsreduzierende Verfahren sowie Scaling, Clustering und Assoziationsanalyse anzuwenden. Für eine konkrete Problemstellung können sie eigenständig ein prädiktives Modell auswählen und anpassen, insbesondere setzen sie flexible Lernstrategien unter Verwendung entsprechender Bausteine (Base Learners, Kernels, Regeln usw.) sowie Hyperparameter-Tuning um.