The path to proton structure at 1% accuracy

NNPDF Collaboration; Stefano Carrazza; Juan Cruz-Martinez; Luigi Del Debbio; Stefano Forte; Tommaso Giani; Shayan Iranipour; Zahari Kassabov; José I. Latorre; Emanuele R. Nocera; Rosalyn L. Pearson; Juan Rojo; Roy Stegeman; Christopher Schwan; Maria Ubiali; Cameron Voisey; Michael Wilson

doi:10.1140/epjc/s10052-022-10328-7

The path to proton structure at 1% accuracy

NNPDF Collaboration
, Stefano Carrazza
, Juan Cruz-Martinez
, Luigi Del Debbio
, Stefano Forte^*
, Tommaso Giani
, Shayan Iranipour
, Zahari Kassabov
, José I. Latorre
, Emanuele R. Nocera
, Rosalyn L. Pearson
, Juan Rojo
, Roy Stegeman
, Christopher Schwan
, Maria Ubiali
, Cameron Voisey
, Michael Wilson

^*Corresponding author for this work

School of Physics and Astronomy

Research output: Contribution to journal › Article › peer-review

Abstract

We present a new set of parton distribution functions (PDFs) based on a fully global dataset and machine learning techniques: NNPDF4.0. We expand the NNPDF3.1
determination with 44 new datasets, mostly from the LHC. We derive a novel methodology through hyperparameter optimization, leading to an efficient fitting algorithm built upon stochastic gradient descent. We use NNLO QCD cal-
culations and account for NLO electroweak corrections and nuclear uncertainties. Theoretical improvements in the PDF description include a systematic implementation of positivity constraints and integrability of sum rules. We validate our methodology by means of closure tests and “future tests” (i.e. tests of backward and forward data compatibility), and assess its stability, specifically upon changes of PDF parametrization basis. We study the internal compatibility of our dataset,
and investigate the dependence of results both upon the choice of input dataset and of fitting methodology. We perform a first study of the phenomenological implications of NNPDF4.0 on representative LHC processes. The software
framework used to produce NNPDF4.0 is made available as an open-source package together with documentation and examples.

Original language	English
Article number	428
Pages (from-to)	1-119
Number of pages	119
Journal	The European Physical Journal C
Volume	82
Issue number	5
DOIs	https://doi.org/10.1140/epjc/s10052-022-10328-7
Publication status	Published - 11 May 2022

Access to Document

10.1140/epjc/s10052-022-10328-7Licence: Creative Commons: Attribution (CC-BY)

Particle Theory at the Higgs Centre
Ball, R. (Principal Investigator), Boyle, P. (Co-investigator), Del Debbio, L. (Co-investigator), Gardi, E. (Co-investigator), Horsley, R. (Co-investigator), Kennedy, A. (Co-investigator), O'Connell, D. (Co-investigator), Smillie, J. (Co-investigator) & Zwicky, R. (Co-investigator)
STFC
1/10/17 → 30/09/21
Project: Research

Cite this

NNPDF Collaboration, Carrazza, S., Cruz-Martinez, J., Del Debbio, L., Forte, S., Giani, T., Iranipour, S., Kassabov, Z., Latorre, J. I., Nocera, E. R., Pearson, R. L., Rojo, J., Stegeman, R., Schwan, C., Ubiali, M., Voisey, C., & Wilson, M. (2022). The path to proton structure at 1% accuracy. The European Physical Journal C, 82(5), 1-119. Article 428. https://doi.org/10.1140/epjc/s10052-022-10328-7

@article{eb9108049c824475a06aab23c9abff92,

title = "The path to proton structure at 1\% accuracy",

abstract = "We present a new set of parton distribution functions (PDFs) based on a fully global dataset and machine learning techniques: NNPDF4.0. We expand the NNPDF3.1 determination with 44 new datasets, mostly from the LHC. We derive a novel methodology through hyperparameter optimization, leading to an efficient fitting algorithm built upon stochastic gradient descent. We use NNLO QCD cal- culations and account for NLO electroweak corrections and nuclear uncertainties. Theoretical improvements in the PDF description include a systematic implementation of positivity constraints and integrability of sum rules. We validate our methodology by means of closure tests and “future tests” (i.e. tests of backward and forward data compatibility), and assess its stability, specifically upon changes of PDF parametrization basis. We study the internal compatibility of our dataset, and investigate the dependence of results both upon the choice of input dataset and of fitting methodology. We perform a first study of the phenomenological implications of NNPDF4.0 on representative LHC processes. The software framework used to produce NNPDF4.0 is made available as an open-source package together with documentation and examples.",

author = "\{NNPDF Collaboration\} and Ball, \{Richard D.\} and Stefano Carrazza and Juan Cruz-Martinez and \{Del Debbio\}, Luigi and Stefano Forte and Tommaso Giani and Shayan Iranipour and Zahari Kassabov and Latorre, \{Jos{\'e} I.\} and Nocera, \{Emanuele R.\} and Pearson, \{Rosalyn L.\} and Juan Rojo and Roy Stegeman and Christopher Schwan and Maria Ubiali and Cameron Voisey and Michael Wilson",

year = "2022",

month = may,

day = "11",

doi = "10.1140/epjc/s10052-022-10328-7",

language = "English",

volume = "82",

pages = "1--119",

journal = "The European Physical Journal C",

issn = "1434-6044",

publisher = "Springer Nature",

number = "5",

}

NNPDF Collaboration, Carrazza, S, Cruz-Martinez, J, Del Debbio, L, Forte, S, Giani, T, Iranipour, S, Kassabov, Z, Latorre, JI, Nocera, ER, Pearson, RL, Rojo, J, Stegeman, R, Schwan, C, Ubiali, M, Voisey, C & Wilson, M 2022, 'The path to proton structure at 1% accuracy', The European Physical Journal C, vol. 82, no. 5, 428, pp. 1-119. https://doi.org/10.1140/epjc/s10052-022-10328-7

TY - JOUR

T1 - The path to proton structure at 1% accuracy

AU - NNPDF Collaboration

AU - Ball, Richard D.

AU - Carrazza, Stefano

AU - Cruz-Martinez, Juan

AU - Del Debbio, Luigi

AU - Forte, Stefano

AU - Giani, Tommaso

AU - Iranipour, Shayan

AU - Kassabov, Zahari

AU - Latorre, José I.

AU - Nocera, Emanuele R.

AU - Pearson, Rosalyn L.

AU - Rojo, Juan

AU - Stegeman, Roy

AU - Schwan, Christopher

AU - Ubiali, Maria

AU - Voisey, Cameron

AU - Wilson, Michael

PY - 2022/5/11

Y1 - 2022/5/11

N2 - We present a new set of parton distribution functions (PDFs) based on a fully global dataset and machine learning techniques: NNPDF4.0. We expand the NNPDF3.1 determination with 44 new datasets, mostly from the LHC. We derive a novel methodology through hyperparameter optimization, leading to an efficient fitting algorithm built upon stochastic gradient descent. We use NNLO QCD cal- culations and account for NLO electroweak corrections and nuclear uncertainties. Theoretical improvements in the PDF description include a systematic implementation of positivity constraints and integrability of sum rules. We validate our methodology by means of closure tests and “future tests” (i.e. tests of backward and forward data compatibility), and assess its stability, specifically upon changes of PDF parametrization basis. We study the internal compatibility of our dataset, and investigate the dependence of results both upon the choice of input dataset and of fitting methodology. We perform a first study of the phenomenological implications of NNPDF4.0 on representative LHC processes. The software framework used to produce NNPDF4.0 is made available as an open-source package together with documentation and examples.

AB - We present a new set of parton distribution functions (PDFs) based on a fully global dataset and machine learning techniques: NNPDF4.0. We expand the NNPDF3.1 determination with 44 new datasets, mostly from the LHC. We derive a novel methodology through hyperparameter optimization, leading to an efficient fitting algorithm built upon stochastic gradient descent. We use NNLO QCD cal- culations and account for NLO electroweak corrections and nuclear uncertainties. Theoretical improvements in the PDF description include a systematic implementation of positivity constraints and integrability of sum rules. We validate our methodology by means of closure tests and “future tests” (i.e. tests of backward and forward data compatibility), and assess its stability, specifically upon changes of PDF parametrization basis. We study the internal compatibility of our dataset, and investigate the dependence of results both upon the choice of input dataset and of fitting methodology. We perform a first study of the phenomenological implications of NNPDF4.0 on representative LHC processes. The software framework used to produce NNPDF4.0 is made available as an open-source package together with documentation and examples.

U2 - 10.1140/epjc/s10052-022-10328-7

DO - 10.1140/epjc/s10052-022-10328-7

M3 - Article

SN - 1434-6044

VL - 82

SP - 1

EP - 119

JO - The European Physical Journal C

JF - The European Physical Journal C

IS - 5

M1 - 428

ER -

The path to proton structure at 1% accuracy

Abstract

Access to Document

Fingerprint

Projects

Particle Theory at the Higgs Centre

Cite this