Today we observe an exploding number of available data and fast progress in information technology. On the one hand more complex models can thus be used for data description and on the other hand new approaches are necessary to profit from the large amount of data. Based on a precise understanding of the underlying probabilistic and statistical mechanisms, we construct new methods especially for high-dimensional data, statistical inverse problems and for stochastic processes. Their mathematical analysis leads to interesting questions at the intersection of mathematical statistics, statistical learning and applied probability theory.
- DFG project TR 1349/3-1 "High-dimensional statistics for point and jump processes"
- Data Science in Hamburg - Helmholtz Graduate School for the Structure of Matter (DASHH)