Research

Paper

TESTING February 27, 2026

A Machine Learning Approach for Lattice Gauge Fixing

Authors

Ho Hsiao, Benjamin J. Choi, Hiroshi Ohno, Akio Tomiya

Abstract

Gauge fixing is an essential step in lattice QCD calculations, particularly for studying gauge-dependent observables. Traditional iterative algorithms are computationally expensive and often suffer from critical slowing down and scaling bottlenecks on large lattices. We present a novel machine learning framework for lattice gauge fixing, where Wilson lines are utilized to construct gauge transformation matrices within a convolutional neural network. The model parameters are optimized via backpropagation, and we introduce a hybrid strategy that combines a neural-network-based transformation with subsequent iterative methods. Preliminary tests on SU(3) gauge theory ensembles for Coulomb gauge demonstrate the potential of this approach to improve the efficiency of lattice gauge fixing. Furthermore, we show that the model exhibits lattice size transferability, where parameters optimized on smaller lattices remain effective for larger volumes without additional training. This framework provides a scalable path toward mitigating critical slowing down in high-precision gauge fixing.

Metadata

arXiv ID: 2602.23731
Provider: ARXIV
Primary Category: hep-lat
Published: 2026-02-27
Fetched: 2026-03-02 06:04

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.23731v1</id>\n    <title>A Machine Learning Approach for Lattice Gauge Fixing</title>\n    <updated>2026-02-27T06:57:20Z</updated>\n    <link href='https://arxiv.org/abs/2602.23731v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.23731v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Gauge fixing is an essential step in lattice QCD calculations, particularly for studying gauge-dependent observables. Traditional iterative algorithms are computationally expensive and often suffer from critical slowing down and scaling bottlenecks on large lattices. We present a novel machine learning framework for lattice gauge fixing, where Wilson lines are utilized to construct gauge transformation matrices within a convolutional neural network. The model parameters are optimized via backpropagation, and we introduce a hybrid strategy that combines a neural-network-based transformation with subsequent iterative methods. Preliminary tests on SU(3) gauge theory ensembles for Coulomb gauge demonstrate the potential of this approach to improve the efficiency of lattice gauge fixing. Furthermore, we show that the model exhibits lattice size transferability, where parameters optimized on smaller lattices remain effective for larger volumes without additional training. This framework provides a scalable path toward mitigating critical slowing down in high-precision gauge fixing.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='hep-lat'/>\n    <published>2026-02-27T06:57:20Z</published>\n    <arxiv:comment>10 pages, 4 figures, 2 tables, Proceedings of the 42nd International Symposium on Lattice Field Theory (Lattice 2025), November 2nd - 8th, 2025, TIFR Mumbai, India</arxiv:comment>\n    <arxiv:primary_category term='hep-lat'/>\n    <author>\n      <name>Ho Hsiao</name>\n    </author>\n    <author>\n      <name>Benjamin J. Choi</name>\n    </author>\n    <author>\n      <name>Hiroshi Ohno</name>\n    </author>\n    <author>\n      <name>Akio Tomiya</name>\n    </author>\n  </entry>"
}