The German Research Foundation (DFG) has approved the new research unit ‘Active Learning for Systems and Control - Data Informativity, Uncertainty, and Guarantees’. In addition to the Institute of Automatic Control (IRT) at LUH, the University of Freiburg and the Technical Universities in Hamburg, Ilmenau and Munich are also involved. The spokesperson for the research unit is Prof. Dr.-Ing. Matthias Müller from IRT.
From autonomous driving to medical diagnosis: machine learning is used in many areas of everyday life. In practice, however, additional challenges arise, for example in the question of safety guarantees in autonomous driving or in the interaction between humans and machines. ‘Such guarantees are generally not available for conventional machine learning methods, so new methods are required,’ says Prof. Dr.-Ing. Matthias Müller. The research unit therefore aims to develop fundamentally new approaches to active learning in which the learning process is continuously influenced. ‘Such active learning strategies are necessary to ensure the safe, high-performance and data-efficient operation of complex and dynamic systems,’ says Müller. To this end, the researchers are investigating what, when and how the system needs to actively learn. The research results can be used in robotics and energy technology, for example.
