Speaker
Amir Farbin
(University of Texas at Arlington)
Description
Graph Neural Networks (GNNs) have emerged as powerful tools for particle reconstruction in high-energy physics experiments, particularly in calorimeters with irregular geometries, such as those used in the ATLAS experiment. In this work, we present a GNN-based approach to reconstruct particle showers, improve energy resolution, spatial localization, and particle identification. We discuss the model architecture, training strategies, and performance benchmarks, demonstrating the advantages of GNNs over conventional techniques. Our findings highlight the potential of GNNs to enhance calorimeter-based event reconstruction, paving the way for more precise measurements in future collider experiments.
| AI keywords | Graph neutral networks; clustering; reconstruction; calorimeter |
|---|
Primary author
Amir Farbin
(University of Texas at Arlington)
Co-authors
Dr
Debottam Bakshi Gupta
(University of Texas at Arlington)
Mr
Kyle Jones
(University of Texas at Arlington)
Mohammadali Ghaznavi
(University of Texas at Arlington)
