250090 SE Applied Machine Learning (2025S)

This (project) seminar provides a forum for students to work out, present and discuss topics of "Applied Machine Learning" and an opportunity to deepen your knowledge obtained from previous machine learning courses such as the same-titled lecture and pro-seminar (250039/250044).

Topics may cover, but are not limited to,
-) computational/numerical aspects, approaches and algorithms related to machine learning including, but not limited to, supervised and unsupervised learning methods, learning algorithms and computational optimization, reinforcement learning, extreme learning, deep learning, transfer learning, kernel methods, high-dimensional (parametric) (differential) equations, physics-informed and physics-aware machine learning, physics-informed neural networks, genetic algorithms, computer vision, evolution-based methods, automated knowledge acquisition, visualization of patterns in data, multi-strategy learning, multi-agent learning.
-) statistical modeling and engineering applications of machine learning / artificial intelligence including, but not limited to, data mining & big data, computer vision, natural language processing (NLP), intelligent systems, neural networks, AI-based software engineering, robotics and control, physics & astronomy, quantum methods in machine learning, bioinformatics, medicine, healthcare, biology, climate, education, business, finance and social sciences.
-) explainable AI (XAI) and trustworthy AI (TAI)

Topics freely chosen by the students can be pursued in consultation with the supervisor. Students are expected to conduct an independent literature search from text book chapter(s), scientific research publication(s), thesis/theses etc.
At least two of the following aspects should be covered: modeling, (numerical) analysis, numerical methods/algorithms, computer simulations, applications.
Each student develops a selected topic for an oral presentation and a written report, under the guidance of the professor but otherwise independently.