The CERN High-Luminosity Large Hadron Collider (HL-LHC) project aims to
significantly increase the number of collisions in the LHC, posing a large data
volume and complexity challenge to the current reconstruction algorithms.

The project is the result of my ongoing internship with the CMS group at CERN,
during which I explore the use of Graph Neural Network (GNN) approaches for the
task of trackster linking with the goal of deploying the solution scalable to
large data volumes within the TICL framework to improve the reconstruction
performance of particle showers in the novel High-Granularity Calorimeter
(HGCAL) designed by the CMS Collaboration to sustain the harsher conditions of
the HL-LHC.

The HGCAL is an endcap detector used for the study of high-energy particle
collisions. One of the key challenges in the analysis of data from the HGCAL is
the task of trackster linking, which are three-dimensional structures formed by
energy deposits in the calorimeter. Tracksters formed by the same particle are
frequently fragmented due to gaps between detector layers, noise, physical
processes, and particle overlaps. The objective of trackster linking is to
reconnect the incomplete fragments generated by the same particle into
well-formed tracksters for a better particle flow recognition.