Towards a data-driven model of hadronization using normalizing flows

被引：0
|
作者：
Bierlich, Christian [1 ]
Ilten, Phil [2 ]
Menzo, Tony [2 ,3 ,4 ]
Mrenna, Stephen [2 ,5 ]
Szewc, Manuel [2 ]
Wilkinson, Michael K. [2 ]
Youssef, Ahmed [2 ]
Zupan, Jure [2 ,3 ,4 ]
机构：
[1] Lund Univ, Dept Phys, Box 118, SE-22100 Lund, Sweden
[2] Univ Cincinnati, Dept Phys, Cincinnati, OH 45221 USA
[3] Univ Calif Berkeley, Berkeley Ctr Theoret Phys, Berkeley, CA 94720 USA
[4] Lawrence Berkeley Natl Lab, Theoret Phys Grp, Berkeley, CA 94720 USA
[5] Fermilab Natl Accelerator Lab, Sci Comp Div, Batavia, IL 60510 USA
来源：
SCIPOST PHYSICS | 2024年 / 17卷 / 02期
关键词：
QCD MODEL;
D O I：
10.21468/SciPostPhys.17.2.045
中图分类号：
O4 [物理学];
学科分类号：
0702 ;
摘要：
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.
引用
收藏
页数：26
相关论文
共 50 条
  • [41] Towards an integrated longitudinal and lateral movement data-driven model for mixed traffic
    Papathanasopoulou, Vasileia
    Antoniou, Constantinos
    21ST EURO WORKING GROUP ON TRANSPORTATION MEETING (EWGT 2018), 2019, 37 : 489 - 496
  • [42] Towards direct data-driven model-free design of optimal controllers
    Selvi, Daniela
    Piga, Dario
    Bemporad, Alberto
    2018 EUROPEAN CONTROL CONFERENCE (ECC), 2018, : 2836 - 2841
  • [43] Data-driven topology design using a deep generative model
    Yamasaki, Shintaro
    Yaji, Kentaro
    Fujita, Kikuo
    STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION, 2021, 64 (03) : 1401 - 1420
  • [44] Data-driven topology design using a deep generative model
    Shintaro Yamasaki
    Kentaro Yaji
    Kikuo Fujita
    Structural and Multidisciplinary Optimization, 2021, 64 : 1401 - 1420
  • [45] Multiscale constitutive model using data-driven yield function
    Park, Hyungbum
    Cho, Maenghyo
    COMPOSITES PART B-ENGINEERING, 2021, 216
  • [46] A model of fake data in data-driven analysis
    Li, Xiaofan
    Whinston, Andrew B.
    JOURNAL OF MACHINE LEARNING RESEARCH, 2020, 21
  • [47] Data-Driven Modeling of Flows of Antalya Basin and Reconstruction of Missing Data
    Fatih Dikbas
    Mutlu Yasar
    Iranian Journal of Science and Technology, Transactions of Civil Engineering, 2020, 44 : 1335 - 1344
  • [48] Data-Driven Modeling of Flows of Antalya Basin and Reconstruction of Missing Data
    Dikbas, Fatih
    Yasar, Mutlu
    IRANIAN JOURNAL OF SCIENCE AND TECHNOLOGY-TRANSACTIONS OF CIVIL ENGINEERING, 2020, 44 (04) : 1335 - 1344
  • [49] Towards integration of data-driven agronomic experiments with data provenance
    Serra da Cruz, Sergio Manuel
    Pires do Nascimento, Jose Antonio
    COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2019, 161 : 14 - 28
  • [50] Towards a Data Engineering Process in Data-Driven Systems Engineering
    Petersen, Patrick
    Stage, Hanno
    Langner, Jacob
    Ries, Lennart
    Rigoll, Philipp
    Hohl, Carl Philipp
    Sax, Eric
    2022 IEEE INTERNATIONAL SYMPOSIUM ON SYSTEMS ENGINEERING (ISSE), 2022,
12345