We investigate a neural-network-based hypothesis test to distinguish different and charged scalar resonances through the channel at hadron colliders. This is traditionally challenging due to a fourfold ambiguity at proton-proton colliders, such as the Large Hadron Collider. Of the neural network approaches we study, we find a multiclass classifier based on a fully connected neural network trained upon two-dimensional histograms made from kinematic variables of the final state to be the most powerful. Furthermore, by considering the one-jet processes, we demonstrate that one can generalize to multiple two-dimensional histograms to represent different variable pairs. Finally, as a comparison to traditional approaches, we compare our method with Bayesian hypothesis testing and discuss the pros and cons of each approach. The neural network scheme presented in this paper is a powerful tool that can help probe the properties of charged resonances.
{ "_oai": { "updated": "2022-04-05T09:11:16Z", "id": "oai:repo.scoap3.org:60297", "sets": [ "PRD" ] }, "authors": [ { "raw_name": "Spencer Chang", "affiliations": [ { "country": "USA", "value": "Department of Physics and Institute for Fundamental Science University of Oregon, Eugene, Oregon 97403, USA" } ], "surname": "Chang", "given_names": "Spencer", "full_name": "Chang, Spencer" }, { "raw_name": "Ting-Kuo Chen", "affiliations": [ { "country": "Taiwan", "value": "Department of Physics, National Taiwan University, Taipei 10617, Taiwan" } ], "surname": "Chen", "given_names": "Ting-Kuo", "full_name": "Chen, Ting-Kuo" }, { "raw_name": "Cheng-Wei Chiang", "affiliations": [ { "country": "Taiwan", "value": "Department of Physics, National Taiwan University, Taipei 10617, Taiwan" }, { "country": "China", "value": "Physics Division, National Center for Theoretical Sciences, Taipei, Taiwan 10617, Republic of China" } ], "surname": "Chiang", "given_names": "Cheng-Wei", "full_name": "Chiang, Cheng-Wei" } ], "titles": [ { "source": "APS", "title": "Distinguishing <math><msup><mi>W</mi><mo>\u2032</mo></msup></math> signals at hadron colliders using neural networks" } ], "dois": [ { "value": "10.1103/PhysRevD.103.036016" } ], "publication_info": [ { "journal_volume": "103", "journal_title": "Physical Review D", "material": "article", "journal_issue": "3", "year": 2021 } ], "$schema": "http://repo.scoap3.org/schemas/hep.json", "acquisition_source": { "date": "2022-04-05T09:04:45.891778", "source": "APS", "method": "APS", "submission_number": "5f41244cb4bf11ec837fd6d834be26e1" }, "page_nr": [ 24 ], "license": [ { "url": "https://creativecommons.org/licenses/by/4.0/", "license": "CC-BY-4.0" } ], "copyright": [ { "statement": "Published by the American Physical Society", "year": "2021" } ], "control_number": "60297", "record_creation_date": "2021-02-24T17:30:03.696270", "_files": [ { "checksum": "md5:75429acf83216cef7e6ef4c5c5e6ef11", "filetype": "pdf", "bucket": "6f36640b-7c7e-4175-9052-80cb31052ac0", "version_id": "64febe7a-9668-4ae2-a0aa-f275efc56c5f", "key": "10.1103/PhysRevD.103.036016.pdf", "size": 2803348 }, { "checksum": "md5:a45fa5062293579ff1490e821da95a12", "filetype": "xml", "bucket": "6f36640b-7c7e-4175-9052-80cb31052ac0", "version_id": "f38841a5-b4df-4604-ba28-987892c1f2d8", "key": "10.1103/PhysRevD.103.036016.xml", "size": 240127 } ], "collections": [ { "primary": "HEP" }, { "primary": "Citeable" }, { "primary": "Published" } ], "arxiv_eprints": [ { "categories": [ "hep-ph" ], "value": "2007.14586" } ], "abstracts": [ { "source": "APS", "value": "We investigate a neural-network-based hypothesis test to distinguish different <math><msup><mi>W</mi><mo>\u2032</mo></msup></math> and charged scalar resonances through the <math><mrow><mo>\u2113</mo><mo>+</mo><msub><mrow><mrow><mi>E</mi></mrow></mrow><mrow><mi>T</mi></mrow></msub></mrow></math> channel at hadron colliders. This is traditionally challenging due to a fourfold ambiguity at proton-proton colliders, such as the Large Hadron Collider. Of the neural network approaches we study, we find a multiclass classifier based on a fully connected neural network trained upon two-dimensional histograms made from kinematic variables of the final state <math><mo>\u2113</mo></math> to be the most powerful. Furthermore, by considering the one-jet processes, we demonstrate that one can generalize to multiple two-dimensional histograms to represent different variable pairs. Finally, as a comparison to traditional approaches, we compare our method with Bayesian hypothesis testing and discuss the pros and cons of each approach. The neural network scheme presented in this paper is a powerful tool that can help probe the properties of charged resonances." } ], "imprints": [ { "date": "2021-02-24", "publisher": "APS" } ] }