Research

Our research brings together fundamental methods, clinical questions, and the systems needed to develop robust, frugal and embedded artificial intelligence.

AI Foundations

Machine learning, generative modeling and computer vision, with an emphasis on Bayesian inference, uncertainty quantification, representation learning and robustness.

AI for Health

Learning from limited, heterogeneous or imperfect medical data to support diagnosis, prognosis and clinical decisions, with a two-way exchange between clinical practice and algorithmic innovation.

Systems and Knowledge Management for AI

Efficient and explainable data and computing infrastructures, privacy-preserving federated learning, resource-aware language models, knowledge graphs and symbolic reasoning.

Across these pillars, we promote open and reproducible research and connect scientific methods with applications in health, digital infrastructure, creative industries and intelligent mobility.