Speaker
Description
High-granularity electromagnetic calorimeters play a crucial role in photon identification in high-energy particle and nuclear physics experiments. A major challenge is distinguishing isolated photons ($\gamma$) from highly boosted neutral pions ($\pi^0$), whose decay photons can produce overlapping electromagnetic showers that mimic a single-photon signal. The fine spatial segmentation of modern calorimeters enables detailed characterization of transverse shower profiles, providing an effective approach to $\gamma$\$\pi^0$ discrimination.
In this study, we investigate transverse shower-shape observables using a Geant4 simulation of a highly granular silicon–tungsten sampling calorimeter similar to the ALICE Forward Calorimeter (FoCal-E) design. Single-photon and ($\pi^0\ \rightarrow \gamma\gamma$) events at different particle energy are analyzed using transverse shower profiles, lateral shower widths, and energy-weighted spatial distributions. A Boosted Decision Tree (BDT) classifier trained on these observables demonstrats reasonable good discrimination power using shower-shape information alone.
We further plan to extend this study by implementing advanced machine learning architectures, including Graph Neural Networks (GNNs) and Particle Cloud Networks (ParticleNet), to exploit pixel-level spatial correlations and longitudinal shower development without relying exclusively on handcrafted observables. These studies will help evaluate the potential of modern machine learning techniques for $\gamma$\$\pi^0$ discrimination and provide valuable insights for the development and optimization of high-granularity electromagnetic calorimeter systems for future experiments, particularly at the Electron–Ion Collider (EIC).