At MBD, we explore a question that’s more relevant than ever in today’s data-driven world: How can we design innovative, reliable, and generalizable methods to handle massive datasets and solve complex problems?

Our lab, headed by Prof. Dr. Nadja Klein, specializes in Bayesian learning methods, a powerful approach that allows us to incorporate prior knowledge into models, quantify uncertainties, and bring more clarity to the “black boxes” of machine learning. For example, we leverage expert insights or sparsity-inducing mechanisms to make models more accurate, robust, and data-efficient. By fusing the precision and reliability of Bayesian Statistics with the adaptability of Machine and Deep Learning, we aim to deliver the best of both worlds.

Our research spans theoretical analysis, method development and real-world applications. For instance, some of our members craft new priors, others develop scalable and trustworthy Bayesian neural networks, and some advance explainability of complex systems. On the application side, our methods include diverse fields—from analyzing complex biomedical data and predicting weather patterns to improving autonomous driving technologies.

Further details about the our research activities can be found on our website.

Further information on the research team, advertised HiWi positions and Master's theses can be found under Organization Methods for Big Data.