Machine learning is currently one of the driving forces of innovation. In our research group, we work on research questions that aim at making machine learning applicable to a broader range of applications, e.g. when little supervised data is available, or when the robustness of the resulting methods is important, too.
For this, we rely on techniques from statistical learning theory as well as optimization and modeling. Many of our results can also be tested experimentally, and typically we do so on computer vision problems, such as image classification, object detection and semantic segmentation.