The team of Jan Kieseler leads the consortium and develops advanced machine-learning methods for particle-physics detectors at colliders, translating sensor signals into physics information. This includes graph-based reconstruction, end-to-end optimisation, and scalable inference.
The group of Nicolas Gauger provides mathematical foundations for BRAID, with expertise in optimisation, algorithmic differentiation, and efficient training strategies. Their methods enable differentiable, modular pipelines that connect reconstruction algorithms with complex system designs.
The group of Christian Glaser brings expertise in astroparticle physics, focusing on reconstructing high-energy neutrinos and cosmic rays. They develop modern simulation and machine-learning tools for large observatories such as IceCube-Gen2, providing a complementary non-collider environment to test BRAID’s methods.
The team of Anastasios Belias brings experience from nuclear and hadron-physics detectors at the FAIR facility, with a focus on tracking systems and particle identification. Their instrumentation environment provides a realistic testbed for applying and validating BRAID’s reconstruction methods in changing environments.
Our associated partner inomed contributes expertise from medical technology, in particular real-time neuromonitoring during surgery. This provides a non-physics application where BRAID’s reconstruction and signal-processing approaches can support robust analysis of continuous physiological data.