High-granularity detectors offer unique possibilities to resolve the development of particle trajectories and showers, separate particles, and identify them with high precision. There is particular potential in high granularity calorimeters which - as calorimeters - measure the particle energies by stopping them. However, these calorimeters can also be operated as tracking detectors, making it possible to follow the trajectories e.g. of muons passing through or connect individual parts of hadronic showers, provided the energy thresholds are sufficiently low.
Analytic solutions based on simplified models and few tuneable parameters might not be able to harness the full potential in particular for dense environments. Machine learning techniques, however, have been very successful in pattern recognition tasks for images or, lately , for point clouds, where algorithmic solutions tend to be overly complex and, at the same time, by far do not reach the same performance.
The seminar will cover applications of neural network architectures to particle reconstruction, in particular shower segmentation, particle identification and energy regression in high granularity calorimeters and touch on applications to multi-particle reconstruction such as particle flow.
The detector geometries and physics considerations call for customised solutions, therefore the focus will be on dedicated developments, such as special graph neural networks and training methods, and how they can be applied to physics problems.