Modul
Machine Learning [M-WIWI-103356]
Credits
9Recurrence
Jedes SemesterDuration
1 SemesterLanguage
German/EnglishLevel
4Version
3Organisation
- KIT-Fakultät für Wirtschaftswissenschaften
Bricks
Identifier | Name | LP |
---|---|---|
T-WIWI-109985 | Project Lab Cognitive Automobiles and Robots | 5 |
T-WIWI-106341 | Machine Learning 2 – Advanced Methods | 4.5 |
T-WIWI-109983 | Project Lab Machine Learning | 5 |
T-WIWI-106340 | Machine Learning 1 - Basic Methods | 4.5 |
Competence Certificate
The module examination is carried out in the form of partial examinations on the selected courses of the module, with which the minimum requirement at creditpoints is fulfilled. The learning control is described in each course. The overall score of the module is made up of the sub-scores weighted with creditpoints and is cut off after the first comma point.
Competence Goal
Students
- Gain knowledge of basic methods in the field of machine learning.
- Understand advanced machine learning concepts and their possible applications.
- Can classify, formally describe and evaluate machine learning methods.
- Can apply their knowledge for the selection of suitable models and methods for selected problems in the field of machine learning.
Prerequisites
None
Content
The topic of machine learning considering real-world challenges of complex application domains is a rapidly expanding field of knowledge and the subject of numerous research and development projects. Large parts of modern AI methods are based on machine-learned models.
The Machine Learning 1 course introduces students to the rapidly evolving field of machine learning by providing a solid foundation that covers the major concepts and techniques in the field. Students will explore various methods of supervised, unsupervised, and reinforcement learning, as well as associated model types ranging from simple linear classifiers to more complex models, such as Deep Neural Networks.
The lecture "Machine Learning 2" covers advanced and modern machine learning methods. Modern learning methods like Self-Supervised-Learning and Contrastive Learning as well as model architectures like Diffusion Models, Transformers, Graph Neural Networks, are introduced.
In the practical courses, scientific tasks in the field of autonomous driving or robotics are solved with modern machine learning methods. There, the techniques of machine learning are practically oriented.
Workload
The total workload for this module is approximately 270 hours.