Unit Heidelberg

(1) un-/semi-/self-supervised learning to tackle lack of training data, 
(2) explainability and uncertainty quantification, 
(3) privacy aware and federated learning, 
(4) sparse predictions from dense inputs and 
(5) interpretable low-dimensional representation learning
(6) representation of scientific knowledge e.g. via application of foundation models
(7) Engagement of AI with the natural world and the natural sciences
(8) Application of AI to health research and the life science

The ELLIS Unit Heidelberg is dedicated to addressing some of the major questions of our time, ranging from understanding the principles of life to addressing the healthcare challenges of the future. Bringing together the unique strengths of EMBL, DKFZ, and Heidelberg University, we form a vibrant ecosystem that drives pioneering breakthroughs and sets new benchmarks in AI for science. Rooted in collaboration, transdisciplinary research, and tailored education, we empower the next generation to solve the grand challenges ahead.