Reconstruction of decays to merged photons using end-to-end deep learning with domain continuation in the CMS detector
(2023) Physical review D — Vol. 108 (2023)
Files
2204.pdf
Open Access - Adobe PDF
- 1.5 MB
Details
- Authors
- Abstract
- A novel technique based on machine learning is introduced to reconstruct the decays of highly Lorentz-boosted particles. Using an end-to-end deep learning strategy, the technique bypasses existing rule-based particle reconstruction methods typically used in high energy physics analyses. It uses minimally processed detector data as input and directly outputs particle properties of interest. The new technique is demonstrated for the reconstruction of the invariant mass of particles decaying in the CMS detector. The decay of a hypothetical scalar particle <math display="inline"><mi mathvariant="script">A</mi></math> into two photons, <math display="inline"><mi mathvariant="script">A</mi><mo stretchy="false">→</mo><mi>γ</mi><mi>γ</mi></math>, is chosen as a benchmark decay. Lorentz boosts <math display="inline"><msub><mi>γ</mi><mi mathvariant="normal">L</mi></msub><mo>=</mo><mn>60</mn><mi>–</mi><mn>600</mn></math> are considered, ranging from regimes where both photons are resolved to those where the photons are closely merged as one object. A training method using domain continuation is introduced, enabling the invariant mass reconstruction of unresolved photon pairs in a novel way. The new technique is validated using <math display="inline"><msup><mi>π</mi><mn>0</mn></msup><mo stretchy="false">→</mo><mi>γ</mi><mi>γ</mi></math> decays in LHC collision data.
- Affiliations
Citations
Benecke, A., Bethani, A., Bruno, G., Bury, F., Caputo, C., David, P., Delaere, C., Donertas, I. S., Giammanco, A., Jaffel, K., Jain, S., Lemaitre, V., Mondal, K., Prisciandaro, J., Taliercio, A., Tran, T. T., Vischia, P., Wertz, S., CMS, & et al. (2023). Reconstruction of decays to merged photons using end-to-end deep learning with domain continuation in the CMS detector. Physical review D, 108. https://doi.org/10.1103/PhysRevD.108.052002 (Original work published 2023)