Speaker
Ho Hsiao
(Center for Computational Sciences, University of Tsukuba)
Description
Gauge fixing is an essential step in lattice QCD calculations, particularly when studying gauge-dependent observables. Traditional iterative algorithms for gauge fixing are computationally expensive and often suffer from critical slowing down near fixed points, as well as scaling bottlenecks on large lattices. We present a novel machine learning framework for lattice gauge fixing, in which Wilson lines are utilised to construct gauge transformation matrices. The training parameters of the model are optimised via gradients obtained from backpropagation of a convolutional neural network. Preliminary tests on SU(3) gauge ensembles demonstrate the potential of this approach to improve the efficiency of lattice gauge fixing.
| Parallel Session (for talks only) | Algorithms and artificial intelligence |
|---|
Authors
Akio Tomiya
(Tokyo Woman’s Christian University)
Benjamin Jaedon Choi
(Center for Computational Sciences, University of Tsukuba)
Dr
Hiroshi Ohno
(Center for Computational Sciences, University of Tsukuba)
Ho Hsiao
(Center for Computational Sciences, University of Tsukuba)
