At the Large Hadron Collider (LHC) at CERN in Geneva, jets are collimated streams of particles originating from the fragmentation of quarks and gluons produced in proton–proton collisions. The relevant physical information about the originating quark or gluon is distributed among dozens of particles measured by the detectors, making its extraction a highly non-trivial task.

The inference of this information is currently a challenging problem that can be effectively addressed using state-of-the-art machine learning techniques. In particular, novel Artificial Intelligence approaches specifically optimized for this task have demonstrated the potential to significantly enhance jet reconstruction performance, thereby enabling new and exciting physics measurements.

The student will be involved in the development and optimization of these AI algorithms. Data from the LHCb and CMS experiments will be used for training, validation, and performance evaluation of the proposed methods.