Potential and limitations of machine-learning approaches to inclusive |Vub| determinations

Biekötter, Anke; Kwok, Ka Wang; Pecjak, Benjamin D.

Abstract

The determination of |Vub| in inclusive semileptonic B → Xuℓν decays will be among the pivotal tasks of Belle II. In this paper we study the potential and limitations of machine-learning approaches that attempt to reduce theory uncertainties by extending the experimentally accessible fiducial region of the B → Xuℓν signal into regions where the B → Xcℓν background is dominant. We find that a deep neural network trained on low-level single particle features offers modest improvement in separating signal from background, compared to BDT set-ups using physicist-engineered high-level features. We further illustrate that while the signal acceptance of such a deep neural network deteriorates in kinematic regions where the signal is small, such as at high hadronic invariant mass, neural networks which exclude kinematic features are flatter in kinematics but less inclusive in the sampling of exclusive hadronic final states at fixed kinematics. The trade-off between these two set-ups is somewhat Monte Carlo dependent, and we study this issue using the multipurpose event generator SHERPA in addition to the widely used B-physics tool EVTGEN.

Citation

Biekötter, A., Kwok, K. W., & Pecjak, B. D. (2022). Potential and limitations of machine-learning approaches to inclusive |Vub| determinations. Journal of High Energy Physics, 2022(143), Article 143. https://doi.org/10.1007/jhep01%282022%29143

• Journal Article Type: Article
• Acceptance Date: Jan 11, 2022
• Online Publication Date: Jan 26, 2022
• Publication Date: 2022-01
• Deposit Date: Feb 16, 2022
• Publicly Available Date: Feb 17, 2022
• Journal: Journal of High Energy Physics
• Print ISSN: 1126-6708
• Electronic ISSN: 1029-8479
• Publisher: Scuola Internazionale Superiore di Studi Avanzati (SISSA)
• Peer Reviewed: Peer Reviewed
• Volume: 2022
• Issue: 143
• Article Number: 143
• DOI: https://doi.org/10.1007/jhep01%282022%29143
• Public URL: https://durham-repository.worktribe.com/output/1216872