{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T03:38:03Z","timestamp":1774409883552,"version":"3.50.1"},"reference-count":87,"publisher":"IOP Publishing","issue":"1","license":[{"start":{"date-parts":[[2025,3,7]],"date-time":"2025-03-07T00:00:00Z","timestamp":1741305600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,3,7]],"date-time":"2025-03-07T00:00:00Z","timestamp":1741305600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100000271","name":"Science and Technology Facilities Council","doi-asserted-by":"crossref","award":["ST\/X003167\/1"],"award-info":[{"award-number":["ST\/X003167\/1"]}],"id":[{"id":"10.13039\/501100000271","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":[[2025,3,31]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>The Matrix-Element Method (MEM) has long been a cornerstone of data analysis in high-energy physics. It leverages theoretical knowledge of parton-level processes and symmetries to evaluate the likelihood of observed events. In parallel, the advent of geometric deep learning has enabled neural network architectures that incorporate known symmetries directly into their design, leading to more efficient learning. This paper presents a novel approach that combines MEM-inspired symmetry considerations with equivariant neural network design for particle physics analysis. Even though Lorentz invariance and permutation invariance over all reconstructed objects are the largest and most natural symmetry in the input domain, we find that they are sub-optimal in most practical search scenarios. We propose a longitudinal boost-equivariant message-passing neural network architecture that preserves relevant discrete symmetries. We present numerical studies demonstrating MEM-inspired architectures achieve new state-of-the-art performance in distinguishing di-Higgs decays to four bottom quarks from the QCD background, with enhanced sample and parameter efficiencies. This synergy between MEM and equivariant deep learning opens new directions for physics-informed architecture design, promising more powerful tools for probing physics beyond the Standard Model.<\/jats:p>","DOI":"10.1088\/2632-2153\/adbab1","type":"journal-article","created":{"date-parts":[[2025,3,7]],"date-time":"2025-03-07T12:28:36Z","timestamp":1741350516000},"page":"015059","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":6,"title":["Optimal equivariant architectures from the symmetries of matrix-element likelihoods"],"prefix":"10.1088","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0414-9497","authenticated-orcid":true,"given":"Daniel","family":"Ma\u00eetre","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7143-715X","authenticated-orcid":true,"given":"Vishal S","family":"Ngairangbam","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8362-0576","authenticated-orcid":false,"given":"Michael","family":"Spannowsky","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2025,3,7]]},"reference":[{"key":"mlstadbab1bib1","doi-asserted-by":"publisher","first-page":"4126","DOI":"10.1143\/JPSJ.57.4126","article-title":"Dynamical likelihood method for reconstruction of events with missing momentum. 1: method and toy models","volume":"57","author":"Kondo","year":"1988","journal-title":"J. Phys. Soc. Japan"},{"key":"mlstadbab1bib2","doi-asserted-by":"publisher","first-page":"836","DOI":"10.1143\/JPSJ.60.836","article-title":"Dynamical likelihood method for reconstruction of events with missing momentum. 2: Mass spectra for 2 \u2014 >2 processes","volume":"60","author":"Kondo","year":"1991","journal-title":"J. Phys. Soc. Japan"},{"key":"mlstadbab1bib3","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.91.112003","article-title":"Precision measurement of the top-quark mass in lepton+jets final states","volume":"91","author":"D0 collaboration","year":"2015","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib4","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.83.074010","article-title":"The matrix element method and QCD radiation","volume":"83","author":"Alwall","year":"2011","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib5","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.84.074002","article-title":"Finding physics signals with shower deconstruction","volume":"84","author":"Soper","year":"2011","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib6","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.87.015019","article-title":"Extracting precise Higgs couplings by using the matrix element method","volume":"87","author":"Andersen","year":"2013","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib7","doi-asserted-by":"publisher","first-page":"JHEP11(2012)043","DOI":"10.1007\/JHEP11(2012)043","article-title":"The matrix element method at next-to-leading order","author":"Campbell","year":"2012","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib8","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.physletb.2015.02.020","article-title":"Discoveries far from the lamppost with matrix elements and ranking","volume":"743","author":"Debnath","year":"2015","journal-title":"Phys. Lett. B"},{"key":"mlstadbab1bib9","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.89.094005","article-title":"Finding physics signals with event deconstruction","volume":"89","author":"Soper","year":"2014","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib10","doi-asserted-by":"publisher","first-page":"JHEP09(2015)083","DOI":"10.1007\/JHEP09(2015)083","article-title":"Extending the Matrix Element Method beyond the Born approximation: Calculating event weights at next-to-leading order accuracy","author":"Martini","year":"2015","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib11","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.95.034001","article-title":"Quark-Gluon tagging with shower deconstruction: unearthing dark matter and Higgs couplings","volume":"95","author":"Ferreira de Lima","year":"2017","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib12","doi-asserted-by":"publisher","first-page":"546","DOI":"10.1140\/epjc\/s10052-019-7030-y","article-title":"HYTREES: combining matrix elements and parton shower for hypothesis testing","volume":"79","author":"Prestel","year":"2019","journal-title":"Eur. Phys. J. C"},{"key":"mlstadbab1bib13","doi-asserted-by":"publisher","first-page":"673","DOI":"10.22323\/1.364.0673","article-title":"The matrix element method as a tool for precision and accuracy","author":"Martini","year":"2020","journal-title":"PoS EPS-HEP2019 P"},{"key":"mlstadbab1bib14","doi-asserted-by":"publisher","first-page":"JHEP04(2021)020","DOI":"10.1007\/JHEP04(2021)020","article-title":"Matrix element regression with deep neural networks-Breaking the CPU barrier","author":"Bury","year":"2021","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib15","doi-asserted-by":"publisher","first-page":"094","DOI":"10.21468\/SciPostPhys.15.3.094","article-title":"Two invertible networks for the matrix element method","volume":"15","author":"Butter","year":"2023","journal-title":"SciPost Phys."},{"key":"mlstadbab1bib16","doi-asserted-by":"publisher","first-page":"759","DOI":"10.1140\/epjc\/s10052-023-11931-y","article-title":"Amplitude-assisted tagging of longitudinally polarised bosons using wide neural networks","volume":"83","author":"Grossi","year":"2023","journal-title":"Eur. Phys. J. C"},{"key":"mlstadbab1bib17","article-title":"Precision-machine learning for the matrix element method","author":"Heimel","year":"2023"},{"key":"mlstadbab1bib18","doi-asserted-by":"publisher","first-page":"JHEP07(2016)069","DOI":"10.1007\/JHEP07(2016)069","article-title":"Jet-images-deep learning edn","author":"de Oliveira","year":"2016","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib19","article-title":"Approximating likelihood ratios with calibrated discriminative classifiers","author":"Cranmer","year":"2015"},{"key":"mlstadbab1bib20","doi-asserted-by":"publisher","first-page":"JHEP05(2017)145","DOI":"10.1007\/JHEP05(2017)145","article-title":"Weakly supervised classification in high energy physics","author":"Dery","year":"2017","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib21","doi-asserted-by":"publisher","first-page":"JHEP10(2017)174","DOI":"10.1007\/JHEP10(2017)174","article-title":"Classification without labels: learning from mixed samples in high energy physics","author":"Metodiev","year":"2017","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib22","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.physrep.2019.11.001","article-title":"Jet substructure at the large hadron collider: a review of recent advances in theory and machine learning","volume":"841","author":"Larkoski","year":"2020","journal-title":"Phys. Rept."},{"key":"mlstadbab1bib23","doi-asserted-by":"publisher","first-page":"JHEP01(2019)121","DOI":"10.1007\/JHEP01(2019)121","article-title":"Energy flow networks: deep sets for particle jets","author":"Komiske","year":"2019","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib24","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.98.052004","article-title":"A guide to constraining effective field theories with machine learning","volume":"98","author":"Brehmer","year":"2018","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib25","doi-asserted-by":"publisher","first-page":"161","DOI":"10.1146\/annurev-nucl-101917-021019","article-title":"Deep learning and its application to LHC physics","volume":"68","author":"Guest","year":"2018","journal-title":"Ann. Rev. Nucl. Part. Sci."},{"key":"mlstadbab1bib26","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.101.056019","article-title":"ParticleNet: jet tagging via particle clouds","volume":"101","author":"Qu","year":"2020","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib27","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1007\/s41781-020-0035-2","article-title":"MadMiner: machine learning-based inference for particle physics","volume":"4","author":"Brehmer","year":"2020","journal-title":"Comput. Softw. Big Sci."},{"key":"mlstadbab1bib28","doi-asserted-by":"publisher","first-page":"014","DOI":"10.21468\/SciPostPhys.7.1.014","article-title":"The machine learning landscape of top taggers","volume":"7","author":"Butter","year":"2019","journal-title":"SciPost Phys."},{"key":"mlstadbab1bib29","article-title":"Machine learning in the search for new fundamental physics","author":"Karagiorgi","year":"2021"},{"key":"mlstadbab1bib30","article-title":"Modern machine learning for LHC physicists","author":"Plehn","year":"2022"},{"key":"mlstadbab1bib31","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.108.012001","article-title":"Comparing point cloud strategies for collider event classification","volume":"108","author":"Onyisi","year":"2023","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib32","author":"Calafiura","year":"2022"},{"key":"mlstadbab1bib33","doi-asserted-by":"publisher","first-page":"305","DOI":"10.1038\/s42254-021-00305-6","article-title":"Simulation-based inference in particle physics","volume":"3","author":"Brehmer","year":"2021","journal-title":"Nat. Rev. Phys."},{"key":"mlstadbab1bib34","doi-asserted-by":"publisher","first-page":"JHEP11(2021)066","DOI":"10.1007\/JHEP11(2021)066","article-title":"A factorisation-aware Matrix element emulator","author":"Ma\u00eetre","year":"2021","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib35","doi-asserted-by":"publisher","first-page":"281","DOI":"10.1038\/s42254-023-00569-0","article-title":"Graph neural networks at the Large Hadron Collider","volume":"5","author":"DeZoort","year":"2023","journal-title":"Nat. Rev. Phys."},{"key":"mlstadbab1bib36","doi-asserted-by":"publisher","first-page":"JHEP05(2024)004","DOI":"10.1007\/JHEP05(2024)004","article-title":"Interpretable deep learning models for the inference and classification of LHC data","author":"Ngairangbam","year":"2024","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib37","doi-asserted-by":"publisher","first-page":"JHEP07(2024)245","DOI":"10.1007\/JHEP07(2024)245","article-title":"Equivariant, safe and sensitive-graph networks for new physics","author":"Bhardwaj","year":"2024","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib38","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.105.112008","article-title":"Permutationless many-jet event reconstruction with symmetry preserving attention networks","volume":"105","author":"Fenton","year":"2022","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib39","doi-asserted-by":"publisher","first-page":"178","DOI":"10.21468\/SciPostPhys.12.5.178","article-title":"SPANet: generalized permutationless set assignment for particle physics using symmetry preserving attention","volume":"12","author":"Shmakov","year":"2022","journal-title":"SciPost Phys."},{"key":"mlstadbab1bib40","doi-asserted-by":"publisher","first-page":"139","DOI":"10.1038\/s42005-024-01627-4","article-title":"Reconstruction of unstable heavy particles using deep symmetry-preserving attention networks","volume":"7","author":"Fenton","year":"2024","journal-title":"Commun. Phys."},{"key":"mlstadbab1bib41","doi-asserted-by":"publisher","first-page":"JHEP09(2024)139","DOI":"10.1007\/JHEP09(2024)139","article-title":"Deep learning to improve the sensitivity of Di-Higgs searches in the 4b channel","author":"Chiang","year":"2024","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib42","doi-asserted-by":"publisher","first-page":"JHEP03(2023)221","DOI":"10.1007\/JHEP03(2023)221","article-title":"Multi-variable integration with a neural network","author":"Ma\u00eetre","year":"2023","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib43","doi-asserted-by":"publisher","first-page":"JHEP02(2024)136","DOI":"10.1007\/JHEP02(2024)136","article-title":"Learning likelihood ratios with neural network classifiers","author":"Rizvi","year":"2024","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib44","doi-asserted-by":"publisher","first-page":"107","DOI":"10.21468\/SciPostPhys.15.3.107","article-title":"Unweighting multijet event generation using factorisation-aware neural networks","volume":"15","author":"Jan\u00dfen","year":"2023","journal-title":"SciPost Phys."},{"key":"mlstadbab1bib45","article-title":"Advancing tools for simulation-based inference","author":"Bahl","year":"2024"},{"key":"mlstadbab1bib46","doi-asserted-by":"publisher","first-page":"2619","DOI":"10.1140\/epjs\/s11734-024-01306-z","article-title":"Foundations of automatic feature extraction at lhc\u2013point clouds and graphs","volume":"233","author":"Bhardwaj","year":"2024","journal-title":"Eur. Phys. J. Spec. Top."},{"key":"mlstadbab1bib47","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1109\/MSP.2017.2693418","article-title":"Geometric deep learning: going beyond Euclidean data","volume":"34","author":"Bronstein","year":"2017","journal-title":"IEEE Signal Process. Mag."},{"key":"mlstadbab1bib48","first-page":"pp 2990","article-title":"Group equivariant convolutional networks","volume":"vol 48","author":"Cohen","year":"2016"},{"key":"mlstadbab1bib49","first-page":"pp 2747","article-title":"On the generalization of equivariance and convolution in neural networks to the action of compact groups","volume":"vol 80","author":"Kondor","year":"2018"},{"key":"mlstadbab1bib50","first-page":"pp 1321","article-title":"Gauge equivariant convolutional networks and the icosahedral CNN","volume":"vol 97","author":"Cohen","year":"2019"},{"key":"mlstadbab1bib51","article-title":"Geometric deep learning: grids, groups, graphs, geodesics, and gauges","author":"Bronstein","year":"2021"},{"key":"mlstadbab1bib52","first-page":"pp 9323","article-title":"E(n) equivariant graph neural networks","volume":"vol 139","author":"Satorras","year":"2021"},{"key":"mlstadbab1bib53","first-page":"p 2021","article-title":"Scalars are universal: equivariant machine learning, structured like classical physics","author":"Villar","year":"2021"},{"key":"mlstadbab1bib54","first-page":"pp 35472","article-title":"Geometric algebra transformer","volume":"vol 36","author":"Brehmer","year":"2023"},{"key":"mlstadbab1bib55","first-page":"pp 992","article-title":"Lorentz group equivariant neural network for particle physics","volume":"vol 119","author":"Bogatskiy","year":"2020"},{"key":"mlstadbab1bib56","article-title":"Symmetry group equivariant architectures for physics","author":"Bogatskiy","year":"2022"},{"key":"mlstadbab1bib57","doi-asserted-by":"publisher","first-page":"JHEP07(2022)030","DOI":"10.1007\/JHEP07(2022)030","article-title":"An efficient Lorentz equivariant graph neural network for jet tagging","author":"Gong","year":"2022","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib58","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.109.056003","article-title":"Does Lorentz-symmetric design boost network performance in jet physics?","volume":"109","author":"Li","year":"2024","journal-title":"Phys. Rev. D"},{"key":"mlstadbab1bib59","doi-asserted-by":"publisher","first-page":"JHEP03(2024)113","DOI":"10.1007\/JHEP03(2024)113","article-title":"Explainable equivariant neural networks for particle physics: PELICAN","author":"Bogatskiy","year":"2024","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib60","doi-asserted-by":"publisher","first-page":"485","DOI":"10.1140\/epjc\/s10052-023-11633-5","article-title":"Lorentz group equivariant autoencoders","volume":"83","author":"Hao","year":"2023","journal-title":"Eur. Phys. J. C"},{"key":"mlstadbab1bib61","article-title":"Equivariance is not all you need: characterizing the utility of equivariant graph neural networks for particle physics tasks","author":"Thais","year":"2023"},{"key":"mlstadbab1bib62","article-title":"Lorentz-equivariant geometric algebra transformers for high-energy physics","author":"Spinner","year":"2024"},{"key":"mlstadbab1bib63","article-title":"Towards an understanding of inductive biases for signal searches at LHC","author":"Ma\u00eetre"},{"key":"mlstadbab1bib64","doi-asserted-by":"publisher","first-page":"289","DOI":"10.1098\/rsta.1933.0009","article-title":"On the problem of the most efficient tests of statistical hypotheses","volume":"231","author":"Neyman","year":"1933","journal-title":"Phil. Trans. R. Soc. A"},{"key":"mlstadbab1bib65","article-title":"Optimal symmetries in binary classification","author":"Ngairangbam","year":"204"},{"key":"mlstadbab1bib66","article-title":"Attention is all you need","volume":"vol 30","author":"Vaswani","year":"2017"},{"key":"mlstadbab1bib67","first-page":"pp 18281","article-title":"Particle transformer for jet tagging","volume":"vol 162","author":"Qu","year":"2022"},{"key":"mlstadbab1bib68","doi-asserted-by":"publisher","first-page":"149","DOI":"10.2307\/1968551","article-title":"On unitary representations of the inhomogeneous Lorentz group","volume":"40","author":"Wigner","year":"1939","journal-title":"Ann. Math."},{"key":"mlstadbab1bib69","article-title":"Deep sets","volume":"vol 30","author":"Zaheer","year":"2017"},{"key":"mlstadbab1bib70","first-page":"pp 6487","article-title":"On the limitations of representing functions on sets","volume":"vol 97","author":"Wagstaff","year":"2019"},{"key":"mlstadbab1bib71","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.99.111601","article-title":"Breakdown of the narrow width approximation for new physics","volume":"99","author":"Berdine","year":"2007","journal-title":"Phys. Rev. Lett."},{"key":"mlstadbab1bib72","doi-asserted-by":"publisher","first-page":"1553","DOI":"10.1109\/TNNLS.2013.2293637","article-title":"On the complexity of neural network classifiers: a comparison between shallow and deep architectures","volume":"25","author":"Bianchini","year":"2014","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"mlstadbab1bib73","first-page":"1","article-title":"Topology of deep neural networks","volume":"21","author":"Naitzat","year":"2020","journal-title":"J. Mach. Learn. Res."},{"key":"mlstadbab1bib74","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.10952296","article-title":"Dataset for \u2018Deep Learning to Improve the Sensitivity of Di-Higgs Searches in the 4b Channel","author":"Hsieh","year":"2024"},{"key":"mlstadbab1bib75","doi-asserted-by":"publisher","first-page":"JHEP07(2014)079","DOI":"10.1007\/JHEP07(2014)079","article-title":"The automated computation of tree-level and next-to-leading order differential cross sections and their matching to parton shower simulations","author":"Alwall","year":"2014","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib76","doi-asserted-by":"publisher","first-page":"8","DOI":"10.21468\/SciPostPhysCodeb.8","article-title":"A comprehensive guide to the physics and usage of PYTHIA 8.3","volume":"2022","author":"Bierlich","year":"2022","journal-title":"SciPost Phys. Codeb."},{"key":"mlstadbab1bib77","doi-asserted-by":"publisher","first-page":"JHEP02(2014)057","DOI":"10.1007\/JHEP02(2014)057","article-title":"A modular framework for fast simulation of a generic collider experiment","author":"DELPHES 3 collaboration and DELPHES 3","year":"2014","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib78","doi-asserted-by":"publisher","first-page":"1896","DOI":"10.1140\/epjc\/s10052-012-1896-2","article-title":"FastJet user manual","volume":"72","author":"Cacciari","year":"2012","journal-title":"Eur. Phys. J. C"},{"key":"mlstadbab1bib79","doi-asserted-by":"publisher","first-page":"JHEP04(2008)063","DOI":"10.1088\/1126-6708\/2008\/04\/063","article-title":"The anti-kt jet clustering algorithm","author":"Cacciari","year":"2008","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib80","article-title":"Optimisation of the ATLAS b-tagging performance for the 2016 LHC Run","author":"ATLAS collaboration","year":"2016"},{"key":"mlstadbab1bib81","doi-asserted-by":"publisher","DOI":"10.1088\/1748-0221\/11\/04\/P04008","article-title":"Performance of b-jet identification in the ATLAS experiment","volume":"11","author":"ATLAS collaboration","year":"2016","journal-title":"JINST"},{"key":"mlstadbab1bib82","doi-asserted-by":"publisher","first-page":"681","DOI":"10.1140\/epjc\/s10052-023-11699-1","article-title":"ATLAS flavour-tagging algorithms for the LHC Run 2 pp collision dataset","volume":"83","author":"ATLAS collaboration","year":"2023","journal-title":"Eur. Phys. J. C"},{"key":"mlstadbab1bib83","doi-asserted-by":"publisher","first-page":"JHEP01(2019)030","DOI":"10.1007\/JHEP01(2019)030","article-title":"Search for pair production of Higgs bosons in the bb\u00afbb\u00af final state using proton-proton collisions at s=13 TeV with the ATLAS detector","author":"ATLAS collaboration","year":"2019","journal-title":"J. High Energy Phys."},{"key":"mlstadbab1bib84","article-title":"Fast graph representation learning with PyTorch Geometric","author":"Fey","year":"2019"},{"key":"mlstadbab1bib85","author":"Paszke","year":"2019"},{"key":"mlstadbab1bib86","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3326362","article-title":"Dynamic graph cnn for learning on point clouds","volume":"38","author":"Wang","year":"2019","journal-title":"ACM Trans. Graph."},{"key":"mlstadbab1bib87","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2015"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adbab1","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adbab1\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adbab1\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adbab1\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,7]],"date-time":"2025-03-07T12:28:41Z","timestamp":1741350521000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adbab1"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,7]]},"references-count":87,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,3,7]]},"published-print":{"date-parts":[[2025,3,31]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/adbab1","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,7]]},"assertion":[{"value":"Optimal equivariant architectures from the symmetries of matrix-element likelihoods","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":"2024-11-15","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-02-26","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-03-07","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}