• Supervised, unsupervised, and reinforcement learning approaches
• Regression and classification learning problems
• Estimation of functional dependencies via regression techniques
• Data clustering
• Support vector machines, large margin classification
• Decision tree learning

• Introduction to neural networks and deep earning
• Network architectures
• Neural network training
• Alternative training methods
• Further network architectures

• Use case evaluation
• Model-centric evaluation
• Business-centric evaluation
• Monitoring
• Avoiding common fallacies
• Change management

• Introduction to reinforcement learning
• Markov chains
• Bandit
• Q-Learning
• Reinforcement learning approaches

In this module, students will reflect on current developments in AI. To this end, pertinent topics will be introduced via articles that are then critically evaluated by the students in the form of a written essay.

In the course “Project: AI Use Case”, students choose a project task in accord with their tutor from a variety of options. The goal is to prototypically implement an artificial intelligence model or system in a suitable development environment. The choice of approach, the system or software implemented, and the resulting performance on the task is to be reasoned about, explained, and documented in a project report. To this end, students make practical use of the methodological knowledge acquired in the previous courses by applying them to relevant real-world problems