skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Deep learning in color: towards automated quark/gluon jet discrimination

Abstract

Artificial intelligence offers the potential to automate challenging data-processing tasks in collider physics. Here, to establish its prospects, we explore to what extent deep learning with convolutional neural networks can discriminate quark and gluon jets better than observables designed by physicists. Our approach builds upon the paradigm that a jet can be treated as an image, with intensity given by the local calorimeter deposits. We supplement this construction by adding color to the images, with red, green and blue intensities given by the transverse momentum in charged particles, transverse momentum in neutral particles, and pixel-level charged particle counts. Overall, the deep networks match or outperform traditional jet variables. We also find that, while various simulations produce different quark and gluon jets, the neural networks are surprisingly insensitive to these differences, similar to traditional observables. This suggests that the networks can extract robust physical information from imperfect simulations.

Authors:
 [1];  [1];  [2]
  1. Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States). Center for Theoretical Physics
  2. Harvard Univ., Cambridge, MA (United States). Dept. of Physics
Publication Date:
Research Org.:
Harvard Univ., Cambridge, MA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1360783
Grant/Contract Number:  
SC0013607
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Journal of High Energy Physics (Online)
Additional Journal Information:
Journal Name: Journal of High Energy Physics (Online); Journal Volume: 2017; Journal Issue: 1; Journal ID: ISSN 1029-8479
Publisher:
Springer Berlin
Country of Publication:
United States
Language:
English
Subject:
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS; Jets

Citation Formats

Komiske, Patrick T., Metodiev, Eric M., and Schwartz, Matthew D.. Deep learning in color: towards automated quark/gluon jet discrimination. United States: N. p., 2017. Web. doi:10.1007/JHEP01(2017)110.
Komiske, Patrick T., Metodiev, Eric M., & Schwartz, Matthew D.. Deep learning in color: towards automated quark/gluon jet discrimination. United States. doi:10.1007/JHEP01(2017)110.
Komiske, Patrick T., Metodiev, Eric M., and Schwartz, Matthew D.. Wed . "Deep learning in color: towards automated quark/gluon jet discrimination". United States. doi:10.1007/JHEP01(2017)110. https://www.osti.gov/servlets/purl/1360783.
@article{osti_1360783,
title = {Deep learning in color: towards automated quark/gluon jet discrimination},
author = {Komiske, Patrick T. and Metodiev, Eric M. and Schwartz, Matthew D.},
abstractNote = {Artificial intelligence offers the potential to automate challenging data-processing tasks in collider physics. Here, to establish its prospects, we explore to what extent deep learning with convolutional neural networks can discriminate quark and gluon jets better than observables designed by physicists. Our approach builds upon the paradigm that a jet can be treated as an image, with intensity given by the local calorimeter deposits. We supplement this construction by adding color to the images, with red, green and blue intensities given by the transverse momentum in charged particles, transverse momentum in neutral particles, and pixel-level charged particle counts. Overall, the deep networks match or outperform traditional jet variables. We also find that, while various simulations produce different quark and gluon jets, the neural networks are surprisingly insensitive to these differences, similar to traditional observables. This suggests that the networks can extract robust physical information from imperfect simulations.},
doi = {10.1007/JHEP01(2017)110},
journal = {Journal of High Energy Physics (Online)},
number = 1,
volume = 2017,
place = {United States},
year = {Wed Jan 25 00:00:00 EST 2017},
month = {Wed Jan 25 00:00:00 EST 2017}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 14 works
Citation information provided by
Web of Science

Save / Share: