Contact person: Prof. Dr.-Ing. Ingo Staack
Contact person: Uche Agbogwu
Time: Tuesday, 15:45 - 18:15
Place: HB 35.1 (IFL Seminarraum)
Introduction to machine learning, probability theory, linear regression models, regularisation, extension to Bayesian approaches, dual representation (kernel methods), Gaussian processes (kriging), neural networks, extension to unsupervised learning, sampling, optimisation and efficient numerical methods for Bayesian approaches, graphical models, global perspective of methods via Bayesian statistics.
In this course, students will receive a comprehensive introduction to machine learning techniques and gain the ability to formulate and solve complex probabilistic models using the sum and product rules of probability. Through the machine learning techniques acquired in this course, students will gain the ability to generate models in design optimisation that allow them to automatically and efficiently explore the solutions by exploiting the uncertainties gained in the learning process. In addition, the machine learning techniques learnt in this course can also be used to perform pre-processing such as feature extraction, which is commonly used in image recognition technology. These contribute to problem simplification and cost efficiency in engineering problems in general and also enable automatic pattern generation, i.e. the construction of new images in the example above. In addition, it is used as a key technology in scientific problems to uncover essential physical quantities. Overall, the global view and unification of probability theory from a Bayesian perspective will enable students to actively formulate probabilistic models and acquire appropriate machine learning approaches for any problem. The course includes practical exercises with computer programmes.
Further information and current announcements can be found on Stud.IP