While often studied separately, these fields are complementary: causality uncovers the underlying cause-effect mechanisms of data, while explainability sheds light on the behavior of predictive models. By bringing them together, this will open up new avenues for enhancing the reliability and interpretability of AI models. Designed for Master’s and PhD students, the program combines lectures, tutorials, and interdisciplinary discussions, including an accessible introductory tutorial tailored to Master’s students.

Speakers

• Erwan Scornet, Lecturer in Applied Mathematics - Sorbonne Université, Paris (LPSM, SCAI)

• Gjergji Kasneci, Professor of Responsible Data Science - Technical University of Munich (TUM)

• Grégoire Montavon, Professor at Charité Universitätsmedizin Berlin and Research Group Lead in the Berlin Institute for the Foundations of Learning and Data (BIFOLD)

• Julie Josse, Senior Researcher at Inria (National research center for digital science), leader of the Inria-Inserm team Premedical

• Krikamol Muandet, Chief scientist and tenure-track faculty at CISPA - Helmholtz Center for Information Security

• Mihaela van der Schaar, John Humphrey Plummer Professor of ML, AI & Medicine — University of Cambridge

• Ronan Sicre, Junior Professor Chair - University Toulouse (IRIT, ADRIA)

• Ulrike von Luxburg, Professor of Statistical Learning Theory - University of Tübingen

Organizers

• Damien Garreau, Julius-Maximilians-Universität Würzburg

• Emilie Devijver, CNRS, Université Grenoble Alpes

• Luca Ganassali, Université Paris-Saclay

• Marianne Clausel, Université de Lorraine

• Ricardo Borsoi, CNRS, Université de Lorraine