Towards a data-driven model of hadronization using normalizing flows
We introduce a model of hadronization based on invertible neural networks that faithfully reproduces a simplified version of the Lund string model for meson hadronization. Additionally, we introduce a new training method for normalizing flows, termed MAGIC, that improves the agreement between simulated and experimental distributions of high-level (macroscopic) observables by adjusting single-emission (microscopic) dynamics. Our results constitute an important step toward realizing a machine-learning based model of hadronization that utilizes experimental data during training. Finally, we demonstrate how a Bayesian extension to this normalizing-flow architecture can be used to provide analysis of statistical and modeling uncertainties on the generated observable distributions.
Authors
- P. Ilten (ORCID: https://orcid.org/0000-0001-5534-1732)
- Jure Zupan (ORCID: https://orcid.org/0000-0001-9445-537X)
- Christian Bierlich (ORCID: https://orcid.org/0000-0002-3978-6085)
- Manuel Szewc (ORCID: https://orcid.org/0000-0002-3014-5388)
- M. Wilkinson (ORCID: https://orcid.org/0000-0001-6561-2145)
- Tony Menzo (ORCID: https://orcid.org/0000-0002-2013-4570)
- S. Mrenna (ORCID: https://orcid.org/0000-0001-8731-160X)
- Ahmed Youssef (ORCID: https://orcid.org/0000-0003-1585-4757)
Institutions
- Fermi National Accelerator Laboratory (US)
- Lawrence Berkeley National Laboratory (US)
- Lund University (SE)
- University of Cincinnati (US)
- University of California, Berkeley (US)
Publication Details
- Journal
- SciPost Physics
- Published
- 2024-08-12
- DOI
- https://doi.org/10.21468/scipostphys.17.2.045
- Citations
- 11
- Primary Topic
- High-Energy Particle Collisions Research
- Type
- article
- Field-Weighted Citation Impact
- 7.26
Funders
- National Science Foundation
- U.S. Department of Energy
- Adolph C. and Mary Sprague Miller Institute for Basic Research in Science, University of California Berkeley
- Knut och Alice Wallenbergs Stiftelse
- Office of Science
- Oak Ridge Institute for Science and Education
- High Energy Physics
- Workforce Development for Teachers and Scientists
- Fermilab