Speaker
Description
Graph Neural Network (GNN) has been demonstrated to be a promising technique for particle track reconstruction as it provides better scaling compared to the traditional combinatorial algorithm. Most of the GNN tracking methods can be classified to edge classification and object condensation. A common problem with these approaches is that it does not handle the situation where a spacepoint is shared by multiple particle tracks. In this presentation, we propose a GNN model that learns an embedding space for the spacepoint doublet. A clustering can then be performed in the spacepoint doublet embedding space to extract track candidates. Alternatively, combining with the edge classification approach, the spacepoint doublet embedding can be used to resolve connected components formed by multiple particle tracks sharing the same spacepoints. We take the ATLAS ITk detector at the HL-LHC as a realistic example and show promising tracking efficiency with the ATLAS ITk simulation. We also show that we are able to assign shared spacepoints to multiple track candidates with the learning of edge embedding space, resulting in better track quality.
Significance
This presentation presents the results of a novel approach for particle track reconstruction. The promising track performance shows that this approach can replace or be combined with the existing approaches for track reconstruction.
