Interdisciplinary Approach

For comprehensively tackling small data challenges, we integrate and fuse contributions from computer science, mathematics, statistics/systems modeling, and biomedicine. In particular, this is reflected in the individual projects in the Project Areas A (Similarity), B (Transfer) and C (Uncertainty), often with co-leads from different disciplines. Overarching topics are addressed by the Fusion Hub. Our doctoral program SMART is key for establishing a shared language.

Computer Science

Advances concepts developed in the context of deep learning, such as meta-learning, few-shot learning, and attention mechanisms.


Provides a range of prototypical applications, to ensure broad generalization of the methods developed in their context.


Provides comprehensive mathematical theory, e.g., on parameter estimates, similarity measures, and confidence intervals, for proving the validity of the proposed methodology.

Statistics / Systems Modeling

Addresses knowledge-driven modeling, for incorporating domain knowledge, in particular in combination with data-driven modeling.

Fusion Hub

Coordinates and advances overarching topics in exchange with Methods Cores, representing the different disciplines.

Administrative Manager

Marc Schumacher

Institute of Medical Biometry and Statistics,
Faculty of Medicine and Medical Center –
University of Freiburg