{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,16]],"date-time":"2025-09-16T17:35:20Z","timestamp":1758044120830,"version":"3.44.0"},"reference-count":30,"publisher":"IOP Publishing","issue":"3","license":[{"start":{"date-parts":[[2025,9,12]],"date-time":"2025-09-12T00:00:00Z","timestamp":1757635200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,9,12]],"date-time":"2025-09-12T00:00:00Z","timestamp":1757635200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/100011665","name":"Thomas Jefferson National Accelerator Facility","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100011665","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2025,9,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Nuclear physics experiments are aimed at uncovering the fundamental building blocks of matter. The experiments involve high-energy collisions that produce complex events with many particle trajectories. Tracking charged particles resulting from collisions in the presence of a strong magnetic field is critical to enable the reconstruction of particle trajectories and precise determination of interactions. It is traditionally achieved through combinatorial approaches that scale worse than linearly as the number of hits grows. Since particle hit data naturally form a point cloud and can be structured as graphs, graph neural networks (GNNs) emerge as an intuitive and effective choice for this task. In this study, we evaluate the GNN model for track finding on the data from the GlueX experiment at Jefferson Lab. We use simulation data to train the model and test on both simulation and real GlueX measurements. We demonstrate that GNN-based track finding outperforms the currently used traditional method at GlueX in terms of segment-based efficiency at a fixed purity while providing faster inferences. We show that the GNN model can achieve significant speedup by processing multiple events in batches, which exploits the parallel computation capability of graphical processing units (GPUs). Finally, we compare the GNN implementation on GPU and field-programmable gate array and describe the trade-off.<\/jats:p>","DOI":"10.1088\/2632-2153\/ae02df","type":"journal-article","created":{"date-parts":[[2025,9,3]],"date-time":"2025-09-03T22:52:39Z","timestamp":1756939959000},"page":"035049","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Geometric GNNs for charged particle tracking at GlueX"],"prefix":"10.1088","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6427-3003","authenticated-orcid":true,"given":"Ahmed","family":"Hossam Mohammed","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4430-9937","authenticated-orcid":false,"given":"Kishansingh","family":"Rajput","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0009-0005-2542-9000","authenticated-orcid":false,"given":"Simon","family":"Taylor","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2238-8857","authenticated-orcid":false,"given":"Denis","family":"Furletov","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7178-8929","authenticated-orcid":false,"given":"Sergey","family":"Furletov","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3475-2871","authenticated-orcid":true,"given":"Malachi","family":"Schram","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2025,9,12]]},"reference":[{"key":"mlstae02dfbib1","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1007\/978-3-030-65771-0_2","type":"book","article-title":"Event Reconstruction","author":"Fr\u00fchwirth","year":"2021"},{"key":"mlstae02dfbib2","first-page":"272","type":"book","article-title":"Charged particle reconstruction for future high energy colliders with quantum approximate optimization algorithm","author":"Okawa","year":"2024"},{"article-title":"Combined track finding with GNN & CKF","year":"2024","author":"Heinrich","key":"mlstae02dfbib3","type":"preprint"},{"key":"mlstae02dfbib4","doi-asserted-by":"publisher","DOI":"10.1016\/j.nima.2020.164807","type":"journal-article","volume":"987","author":"Adhikari","year":"2021","journal-title":"Nucl. Instrum. Methods Phys. Res. A"},{"key":"mlstae02dfbib5","doi-asserted-by":"publisher","first-page":"498","DOI":"10.1016\/0168-9002(88)90722-X","type":"journal-article","volume":"270","author":"Hansroul","year":"1988","journal-title":"Nucl. Inst. Methods Phys. Res. A"},{"key":"mlstae02dfbib6","doi-asserted-by":"publisher","first-page":"85","DOI":"10.1016\/j.nima.2006.05.040","type":"journal-article","volume":"566","author":"Kisel","year":"2006","journal-title":"Nucl. Inst. Methods Phys. Res. A"},{"key":"mlstae02dfbib7","doi-asserted-by":"publisher","DOI":"10.1016\/j.nima.2019.163304","type":"journal-article","volume":"956","author":"Brondolin","year":"2020","journal-title":"Nucl. Instrum. Methods Phys. Res. A"},{"key":"mlstae02dfbib8","first-page":"554","type":"journal-article","volume":"590914","author":"Hough","year":"1959","journal-title":"Conf. Proc. C"},{"key":"mlstae02dfbib9","doi-asserted-by":"publisher","DOI":"10.1016\/j.nima.2025.170357","type":"journal-article","volume":"1075","author":"Zhou","year":"2025","journal-title":"2025 Nucl. Instrum. Methods Phys. Res. A"},{"key":"mlstae02dfbib10","doi-asserted-by":"publisher","first-page":"81","DOI":"10.1007\/978-3-030-65771-0_5","type":"journal-article","article-title":"Track Finding","author":"Fr\u00fchwirth","year":"2021","journal-title":"Pattern Recognition, Tracking and Vertex Reconstruction in Particle Detectors"},{"article-title":"Novel deep learning methods for track reconstruction","year":"2018","author":"Farrell","key":"mlstae02dfbib11","type":"preprint"},{"key":"mlstae02dfbib12","first-page":"729","type":"conference-proceedings","article-title":"A new model for learning in graph domains","volume":"vol 2","author":"Gori","year":"2005"},{"key":"mlstae02dfbib13","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1109\/TNN.2008.2005605","type":"journal-article","volume":"20","author":"Scarselli","year":"2009","journal-title":"IEEE Trans. Neural Netw."},{"article-title":"Semi-supervised classification with graph convolutional networks","year":"2017","author":"Kipf","key":"mlstae02dfbib14","type":"preprint"},{"article-title":"Inductive representation learning on large graphs","year":"2018","author":"Hamilton","key":"mlstae02dfbib15","type":"preprint"},{"article-title":"Graph attention networks","year":"2018","author":"Veli\u010dkovi\u0107","key":"mlstae02dfbib16","type":"preprint"},{"article-title":"Graph neural networks for particle reconstruction in high energy physics detectors","year":"2020","author":"Ju","key":"mlstae02dfbib17","type":"preprint"},{"key":"mlstae02dfbib18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1140\/epjc\/s10052-021-09675-8","type":"journal-article","volume":"81","author":"Ju","year":"2021","journal-title":"Eur. Phys. J. C"},{"key":"mlstae02dfbib19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s41781-021-00073-z","type":"journal-article","volume":"5","author":"DeZoort","year":"2021","journal-title":"Comput. Softw. Big Sci."},{"article-title":"Accelerated charged particle tracking with graph neural networks on FPGAs","year":"2020","author":"Heintz","key":"mlstae02dfbib20","type":"preprint"},{"key":"mlstae02dfbib21","doi-asserted-by":"publisher","DOI":"10.3389\/fdata.2022.828666","type":"journal-article","volume":"5","author":"Elabd","year":"2022","journal-title":"Front. Big Data"},{"key":"mlstae02dfbib22","doi-asserted-by":"publisher","first-page":"186","DOI":"10.1016\/j.nima.2016.06.125","type":"journal-article","volume":"835","author":"Allison","year":"2016","journal-title":"Nucl. Instrum. Methods Phys. Res. A"},{"key":"mlstae02dfbib23","first-page":"p 6\/1","type":"conference-proceedings","article-title":"Multiobjective genetic algorithms","author":"Fonseca","year":"1993"},{"key":"mlstae02dfbib24","first-page":"807","type":"conference-proceedings","article-title":"Rectified linear units improve restricted boltzmann machines","author":"Nair","year":"2010"},{"year":"2014","author":"Kingma","key":"mlstae02dfbib25","type":"preprint"},{"year":"2020","author":"Cai","key":"mlstae02dfbib26","type":"preprint"},{"article-title":"A survey on oversmoothing in graph neural networks","year":"2023","author":"Rusch","key":"mlstae02dfbib27","type":"preprint"},{"article-title":"fastmachinelearning\/hls4ml","year":"2024","author":"Team","key":"mlstae02dfbib28","type":"other"},{"article-title":"Vitis High-level synthesis user guide: introduction (UG1399) AMD","year":"2021","author":"AMD","key":"mlstae02dfbib29","type":"web-resource"},{"key":"mlstae02dfbib30","doi-asserted-by":"publisher","first-page":"182","DOI":"10.1109\/4235.996017","type":"journal-article","volume":"6","author":"Deb","year":"2002","journal-title":"IEEE Trans. Evol. Comput."}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae02df","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae02df\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae02df","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae02df\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae02df\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae02df\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae02df\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae02df\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,12]],"date-time":"2025-09-12T10:13:18Z","timestamp":1757671998000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae02df"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,12]]},"references-count":30,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2025,9,12]]},"published-print":{"date-parts":[[2025,9,30]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ae02df","relation":{},"ISSN":["2632-2153"],"issn-type":[{"type":"electronic","value":"2632-2153"}],"subject":[],"published":{"date-parts":[[2025,9,12]]},"assertion":[{"value":"Geometric GNNs for charged particle tracking at GlueX","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2025-05-13","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-09-03","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-09-12","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}