Machine Learning tools for global PDF fits

Juan Rojo

doi:10.48550/arXiv.1809.04392

Machine Learning tools for global PDF fits

Juan Rojo

(Astro)-Particles Physics

Research output: Contribution to Journal › Article › Academic

38 Downloads (Pure)

Abstract

The use of machine learning algorithms in theoretical and experimental high-energy physics has experienced an impressive progress in recent years, with applications from trigger selection to jet substructure classification and detector simulation among many others. In this contribution, we review the machine learning tools used in the NNPDF family of global QCD analyses. These include multi-layer feed-forward neural networks for the model-independent parametrisation of parton distributions and fragmentation functions, genetic and covariance matrix adaptation algorithms for training and optimisation, and closure testing for the systematic validation of the fitting methodology.

Original language	English
Pages (from-to)	1-11
Number of pages	11
Journal	arXiv
Volume	2018
DOIs	https://doi.org/10.48550/arXiv.1809.04392
Publication status	Published - 12 Sept 2018
Event	XXIIIth Quark Confinement and the Hadron Spectrum conference: 1-6 August 2018, University of Maynooth, Ireland - Maynooth, Ireland Duration: 1 Aug 2018 → 6 Aug 2018 Conference number: 23th

Bibliographical note

12 pages, 9 figures, to appear in the proceedings of the XXIIIth Quark Confinement and the Hadron Spectrum conference, 1-6 August 2018, University of Maynooth, Ireland

Keywords

hep-ph

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.48550/arXiv.1809.04392

1809.04392v1Final published version, 1.36 MB

Persistent URL (handle)

Copy link address

Cite this

@article{4a3eaec0372744e3a054fa09ebd685de,

title = "Machine Learning tools for global PDF fits",

abstract = "The use of machine learning algorithms in theoretical and experimental high-energy physics has experienced an impressive progress in recent years, with applications from trigger selection to jet substructure classification and detector simulation among many others. In this contribution, we review the machine learning tools used in the NNPDF family of global QCD analyses. These include multi-layer feed-forward neural networks for the model-independent parametrisation of parton distributions and fragmentation functions, genetic and covariance matrix adaptation algorithms for training and optimisation, and closure testing for the systematic validation of the fitting methodology. ",

keywords = "hep-ph",

author = "Juan Rojo",

note = "12 pages, 9 figures, to appear in the proceedings of the XXIIIth Quark Confinement and the Hadron Spectrum conference, 1-6 August 2018, University of Maynooth, Ireland; XXIIIth Quark Confinement and the Hadron Spectrum conference : 1-6 August 2018, University of Maynooth, Ireland ; Conference date: 01-08-2018 Through 06-08-2018",

year = "2018",

month = sep,

day = "12",

doi = "10.48550/arXiv.1809.04392",

language = "English",

volume = "2018",

pages = "1--11",

journal = "arXiv",

issn = "2331-8422",

publisher = "Cornell University",

}

TY - JOUR

T1 - Machine Learning tools for global PDF fits

AU - Rojo, Juan

N1 - Conference code: 23th

PY - 2018/9/12

Y1 - 2018/9/12

N2 - The use of machine learning algorithms in theoretical and experimental high-energy physics has experienced an impressive progress in recent years, with applications from trigger selection to jet substructure classification and detector simulation among many others. In this contribution, we review the machine learning tools used in the NNPDF family of global QCD analyses. These include multi-layer feed-forward neural networks for the model-independent parametrisation of parton distributions and fragmentation functions, genetic and covariance matrix adaptation algorithms for training and optimisation, and closure testing for the systematic validation of the fitting methodology.

AB - The use of machine learning algorithms in theoretical and experimental high-energy physics has experienced an impressive progress in recent years, with applications from trigger selection to jet substructure classification and detector simulation among many others. In this contribution, we review the machine learning tools used in the NNPDF family of global QCD analyses. These include multi-layer feed-forward neural networks for the model-independent parametrisation of parton distributions and fragmentation functions, genetic and covariance matrix adaptation algorithms for training and optimisation, and closure testing for the systematic validation of the fitting methodology.

KW - hep-ph

UR - https://pos.sissa.it/336/

U2 - 10.48550/arXiv.1809.04392

DO - 10.48550/arXiv.1809.04392

M3 - Article

SN - 2331-8422

VL - 2018

SP - 1

EP - 11

JO - arXiv

JF - arXiv

T2 - XXIIIth Quark Confinement and the Hadron Spectrum conference

Y2 - 1 August 2018 through 6 August 2018

ER -

Machine Learning tools for global PDF fits

Abstract

Bibliographical note

Keywords

UN SDGs

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this