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

Title: New Angle on the Parton Distribution Functions: Self-Organizing Maps

Abstract

Neural network (NN) algorithms have been recently applied to construct Parton Distribution Function (PDF) parametrizations, providing an alternative to standard global fitting procedures. Here we explore a novel technique using Self-Organizing Maps (SOMs). SOMs are a class of clustering algorithms based on competitive learning among spatially-ordered neurons. We train our SOMs with stochastically generated PDF samples. On every optimization iteration the PDFs are clustered on the SOM according to a user-defined feature and the most promising candidates are used as a seed for the subsequent iteration using the topology of the map to guide the PDF generating process. Our goal is a fitting procedure that, at variance with the standard neural network approaches, will allow for an increased control of the systematic bias by enabling user interaction in the various stages of the process.

Authors:
 [1];  [2]
  1. Department of Physics and Astronomy, Iowa State University, Ames, IA 50011 (United States)
  2. Department of Physics, University of Virginia, P.O. Box 400714, Charlottesville, VA 22904-4714 (United States)
Publication Date:
OSTI Identifier:
21316913
Resource Type:
Journal Article
Journal Name:
AIP Conference Proceedings
Additional Journal Information:
Journal Volume: 1149; Journal Issue: 1; Conference: 18. international spin physics symposium, Charlottesville, VA (United States), 6-11 Oct 2008; Other Information: DOI: 10.1063/1.3215649; (c) 2009 American Institute of Physics; Country of input: International Atomic Energy Agency (IAEA); Journal ID: ISSN 0094-243X
Country of Publication:
United States
Language:
English
Subject:
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS; ALGORITHMS; DISTRIBUTION FUNCTIONS; GLUONS; ITERATIVE METHODS; NEURAL NETWORKS; OPTIMIZATION; PARTICLE STRUCTURE; QUANTUM CHROMODYNAMICS; QUARKS; STANDARD MODEL; TOPOLOGY

Citation Formats

Honkanen, H, and Liuti, S. New Angle on the Parton Distribution Functions: Self-Organizing Maps. United States: N. p., 2009. Web. doi:10.1063/1.3215649.
Honkanen, H, & Liuti, S. New Angle on the Parton Distribution Functions: Self-Organizing Maps. United States. https://doi.org/10.1063/1.3215649
Honkanen, H, and Liuti, S. 2009. "New Angle on the Parton Distribution Functions: Self-Organizing Maps". United States. https://doi.org/10.1063/1.3215649.
@article{osti_21316913,
title = {New Angle on the Parton Distribution Functions: Self-Organizing Maps},
author = {Honkanen, H and Liuti, S},
abstractNote = {Neural network (NN) algorithms have been recently applied to construct Parton Distribution Function (PDF) parametrizations, providing an alternative to standard global fitting procedures. Here we explore a novel technique using Self-Organizing Maps (SOMs). SOMs are a class of clustering algorithms based on competitive learning among spatially-ordered neurons. We train our SOMs with stochastically generated PDF samples. On every optimization iteration the PDFs are clustered on the SOM according to a user-defined feature and the most promising candidates are used as a seed for the subsequent iteration using the topology of the map to guide the PDF generating process. Our goal is a fitting procedure that, at variance with the standard neural network approaches, will allow for an increased control of the systematic bias by enabling user interaction in the various stages of the process.},
doi = {10.1063/1.3215649},
url = {https://www.osti.gov/biblio/21316913}, journal = {AIP Conference Proceedings},
issn = {0094-243X},
number = 1,
volume = 1149,
place = {United States},
year = {Tue Aug 04 00:00:00 EDT 2009},
month = {Tue Aug 04 00:00:00 EDT 2009}
}