{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,30]],"date-time":"2026-05-30T04:35:32Z","timestamp":1780115732712,"version":"3.54.0"},"reference-count":45,"publisher":"IOP Publishing","issue":"4","license":[{"start":{"date-parts":[[2023,12,22]],"date-time":"2023-12-22T00:00:00Z","timestamp":1703203200000},"content-version":"vor","delay-in-days":21,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,12,22]],"date-time":"2023-12-22T00:00:00Z","timestamp":1703203200000},"content-version":"tdm","delay-in-days":21,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/100007911","name":"University of California, San Diego","doi-asserted-by":"crossref","award":["TRELS"],"award-info":[{"award-number":["TRELS"]}],"id":[{"id":"10.13039\/100007911","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100006208","name":"High Energy Physics","doi-asserted-by":"crossref","award":["DE-SC0021187"],"award-info":[{"award-number":["DE-SC0021187"]}],"id":[{"id":"10.13039\/100006208","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100000105","name":"Office of Advanced Cyberinfrastructure","doi-asserted-by":"crossref","award":["2117997"],"award-info":[{"award-number":["2117997"]}],"id":[{"id":"10.13039\/100000105","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100006192","name":"Advanced Scientific Computing Research","doi-asserted-by":"crossref","award":["DE-FOA-0002501"],"award-info":[{"award-number":["DE-FOA-0002501"]}],"id":[{"id":"10.13039\/100006192","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":[[2023,12,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>The Earth mover\u2019s distance (EMD) is a useful metric for image recognition and classification, but its usual implementations are not differentiable or too slow to be used as a loss function for training other algorithms via gradient descent. In this paper, we train a convolutional neural network (CNN) to learn a differentiable, fast approximation of the EMD and demonstrate that it can be used as a substitute for computing-intensive EMD implementations. We apply this differentiable approximation in the training of an autoencoder-inspired neural network (encoder NN) for data compression at the high-luminosity LHC at CERN The goal of this encoder NN is to compress the data while preserving the information related to the distribution of energy deposits in particle detectors. We demonstrate that the performance of our encoder NN trained using the differentiable EMD CNN surpasses that of training with loss functions based on mean squared error.<\/jats:p>","DOI":"10.1088\/2632-2153\/ad1139","type":"journal-article","created":{"date-parts":[[2023,11,30]],"date-time":"2023-11-30T22:31:36Z","timestamp":1701383496000},"page":"045058","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":9,"title":["Differentiable Earth mover\u2019s distance for data compression at the high-luminosity LHC"],"prefix":"10.1088","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8738-0218","authenticated-orcid":true,"given":"Rohan","family":"Shenoy","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5076-7096","authenticated-orcid":true,"given":"Javier","family":"Duarte","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4280-6382","authenticated-orcid":false,"given":"Christian","family":"Herwig","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8244-0805","authenticated-orcid":false,"given":"James","family":"Hirschauer","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3932-3769","authenticated-orcid":false,"given":"Daniel","family":"Noonan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1939-4268","authenticated-orcid":true,"given":"Maurizio","family":"Pierini","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8440-6854","authenticated-orcid":false,"given":"Nhan","family":"Tran","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0177-5903","authenticated-orcid":false,"given":"Cristina","family":"Mantilla Suarez","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2023,12,22]]},"reference":[{"key":"mlstad1139bib1","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1023\/A:1026543900054","article-title":"The earth mover\u2019s distance as a metric for image retrieval","volume":"40","author":"Rubner","year":"2000","journal-title":"Int. J. Comput. Vis."},{"key":"mlstad1139bib2","article-title":"Squared earth mover\u2019s distance-based loss for training deep neural networks","author":"Hou","year":"2016"},{"key":"mlstad1139bib3","doi-asserted-by":"publisher","first-page":"274","DOI":"10.1109\/TPAMI.2008.299","article-title":"Differential earth mover\u2019s distance with its applications to visual tracking","volume":"32","author":"Zhao","year":"2010","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"mlstad1139bib4","doi-asserted-by":"publisher","first-page":"621","DOI":"10.1109\/TMI.2010.2089693","article-title":"An optimal transportation approach for nuclear structure-based pathology","volume":"30","author":"Wang","year":"2011","journal-title":"IEEE Trans. Med. Imaging"},{"key":"mlstad1139bib5","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.123.041801","article-title":"Metric space of collider events","volume":"123","author":"Komiske","year":"2019","journal-title":"Phys. Rev. Lett."},{"key":"mlstad1139bib6","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.101.034009","article-title":"Exploring the space of jets with CMS open data","volume":"101","author":"Komiske","year":"2020","journal-title":"Phys. Rev. D"},{"key":"mlstad1139bib7","doi-asserted-by":"publisher","first-page":"JHEP08(2020)084","DOI":"10.1007\/JHEP08(2020)084","article-title":"A robust measure of event isotropy at colliders","author":"Cesarotti","year":"2020","journal-title":"J. High Energy Phys."},{"key":"mlstad1139bib8","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.107.016002","article-title":"Variational autoencoders for anomalous jet tagging","volume":"107","author":"Cheng","year":"2023","journal-title":"Phys. Rev. D"},{"key":"mlstad1139bib9","doi-asserted-by":"publisher","first-page":"192","DOI":"10.1140\/epjc\/s10052-021-08891-6","article-title":"Use of a generalized energy mover\u2019s distance in the search for rare phenomena at colliders","volume":"81","author":"Rom\u00e3o","year":"2021","journal-title":"Eur. Phys. J. C"},{"key":"mlstad1139bib10","article-title":"Particle cloud generation with message passing generative adversarial networks","volume":"vol 34","author":"Kansal","year":"2021"},{"key":"mlstad1139bib11","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR42600.2020.01222","article-title":"DeepEMD: Few-shot image classification with differentiable earth mover\u2019s distance and structured classifiers","author":"Zhang","year":"2020"},{"key":"mlstad1139bib12","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.102.116019","article-title":"Linearized optimal transport for collider events","volume":"102","author":"Cai","year":"2020","journal-title":"Phys. Rev. D"},{"key":"mlstad1139bib13","article-title":"Particle graph autoencoders and differentiable, learned energy Mover\u2019s distance","author":"Tsan","year":"2021"},{"key":"mlstad1139bib14","article-title":"Sinkhorn distances: lightspeed computation of optimal transport","volume":"vol 26","author":"Cuturi","year":"2013"},{"key":"mlstad1139bib15","doi-asserted-by":"crossref","DOI":"10.1007\/JHEP06(2023)195","article-title":"SHAPER: can you hear the shape of a Jet?","author":"Ba","year":"2023"},{"key":"mlstad1139bib16","article-title":"Finding NEEMo: geometric fitting using neural estimation of the energy Mover\u2019s distance","volume":"vol 9","author":"Kitouni","year":"2022"},{"key":"mlstad1139bib17","first-page":"p 11","article-title":"Robust and Provably Monotonic Networks","author":"Kitouni","year":"2021"},{"key":"mlstad1139bib18","doi-asserted-by":"crossref","DOI":"10.1007\/JHEP08(2023)107","article-title":"A spectral metric for collider geometry","author":"Larkoski","year":"2023"},{"key":"mlstad1139bib19","article-title":"Optimal transport for a global event description at high-intensity hadron colliders","author":"Gouskos","year":"2023"},{"key":"mlstad1139bib20","first-page":"p 435","article-title":"Wasserstein barycenter and its application to texture mixing","author":"Rabin","year":"2012"},{"key":"mlstad1139bib21","doi-asserted-by":"publisher","first-page":"2179","DOI":"10.1109\/TNS.2021.3087100","article-title":"A reconfigurable neural network ASIC for detector front-end data compression at the HL-LHC","volume":"68","author":"Di Guglielmo","year":"2021","journal-title":"IEEE Trans. Nucl. Sci."},{"key":"mlstad1139bib22","article-title":"The Phase-2 upgrade of the CMS endcap calorimeter","author":"CMS Collaboration","year":"2017"},{"key":"mlstad1139bib23","article-title":"The Phase-2 upgrade of the CMS Level-1 trigger","author":"CMS Collaboration","year":"2020"},{"key":"mlstad1139bib24","doi-asserted-by":"publisher","DOI":"10.1088\/1748-0221\/15\/10\/P10017","article-title":"Performance of the CMS Level-1 trigger in proton-proton collisions at s= 13 TeV","volume":"15","author":"CMS Collaboration","year":"2020","journal-title":"J. Instrum."},{"key":"mlstad1139bib25","doi-asserted-by":"publisher","DOI":"10.1088\/1748-0221\/15\/10\/P10004","article-title":"Operation of the ATLAS trigger system in Run 2","volume":"15","author":"ATLAS Collaboration","year":"2020","journal-title":"J. Instrum."},{"key":"mlstad1139bib26","article-title":"Technical design report for the phase-II upgrade of the ATLAS TDAQ system","author":"ATLAS Collaboration","year":"2017"},{"key":"mlstad1139bib27","doi-asserted-by":"publisher","first-page":"675","DOI":"10.1038\/s42256-021-00356-5","article-title":"Automatic heterogeneous quantization of deep neural networks for low-latency inference on the edge for particle detectors","volume":"3","author":"Coelho","year":"2021","journal-title":"Nat. Mach. Intell."},{"key":"mlstad1139bib28","first-page":"1","article-title":"POT: python optimal transport","volume":"22","author":"Flamary","year":"2021","journal-title":"J. Mach. Learn. Res."},{"key":"mlstad1139bib29","article-title":"CMS high granularity calorimeter trigger cell simulated dataset (Part 1)","author":"Shenoy","year":"2023"},{"key":"mlstad1139bib30","article-title":"CMS high granularity calorimeter trigger cell simulated dataset (Part 2)","author":"Shenoy","year":"2023"},{"key":"mlstad1139bib31","doi-asserted-by":"publisher","first-page":"250","DOI":"10.1016\/S0168-9002(03)01368-8","article-title":"Geant4\u2014a simulation toolkit","volume":"506","author":"GEANT4 Collaboration","year":"2003","journal-title":"Nucl. Instrum. Methods A"},{"key":"mlstad1139bib32","article-title":"Keras","author":"Chollet","year":"2015"},{"key":"mlstad1139bib33","article-title":"TensorFlow: large-scale machine learning on heterogeneous systems","author":"Abadi","year":"2015"},{"key":"mlstad1139bib34","first-page":"p 448","article-title":"Batch normalization: accelerating deep network training by reducing internal covariate shift","volume":"vol 37","author":"Ioffe","year":"2015"},{"key":"mlstad1139bib35","first-page":"p 807","article-title":"Rectified linear units improve restricted Boltzmann machines","author":"Nair","year":"2010"},{"key":"mlstad1139bib36","first-page":"p 315","article-title":"Deep sparse rectifier neural networks","volume":"vol 15","author":"Glorot","year":"2011"},{"key":"mlstad1139bib37","doi-asserted-by":"publisher","first-page":"73","DOI":"10.1214\/aoms\/1177703732","article-title":"Robust estimation of a location parameter","volume":"35","author":"Huber","year":"1964","journal-title":"Ann. Math. Stat."},{"key":"mlstad1139bib38","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2015"},{"key":"mlstad1139bib39","doi-asserted-by":"publisher","first-page":"192","DOI":"10.1140\/epjc\/s10052-021-08891-6","article-title":"Use of a generalized energy mover\u2019s distance in the search for rare phenomena at colliders","volume":"81","author":"Crispim Rom\u00e3o","year":"2021","journal-title":"Eur. Phys. J. C"},{"key":"mlstad1139bib40","first-page":"p 6","article-title":"Particle cloud generation with message passing generative adversarial networks","author":"Kansal","year":"2021"},{"key":"mlstad1139bib41","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.107.076017","article-title":"Evaluating generative models in high energy physics","volume":"107","author":"Kansal","year":"2023","journal-title":"Phys. Rev. D"},{"key":"mlstad1139bib42","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.106.092009","article-title":"Score-based generative models for calorimeter shower simulation","volume":"106","author":"Mikuni","year":"2022","journal-title":"Phys. Rev. D"},{"key":"mlstad1139bib43","doi-asserted-by":"crossref","DOI":"10.1088\/1748-0221\/18\/11\/P11025","article-title":"CaloClouds: Fast Geometry-Independent Highly-Granular Calorimeter Simulation","author":"Buhmann","year":"2023"},{"key":"mlstad1139bib44","article-title":"Deepemd: a transformer-based fast estimation of the earth mover\u2019s distance","author":"Sinha","year":"2023"},{"key":"mlstad1139bib45","doi-asserted-by":"publisher","DOI":"10.1016\/j.revip.2023.100085","article-title":"Toward the end-to-end optimization of particle physics instruments with differentiable programming","volume":"10","author":"MODE Collaboration","year":"2023","journal-title":"Rev. Phys."}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad1139","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad1139\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad1139","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad1139\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad1139\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad1139\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad1139\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad1139\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,12,22]],"date-time":"2023-12-22T06:55:09Z","timestamp":1703228109000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad1139"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,12,1]]},"references-count":45,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2023,12,22]]},"published-print":{"date-parts":[[2023,12,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ad1139","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,12,1]]},"assertion":[{"value":"Differentiable Earth mover\u2019s distance for data compression at the high-luminosity LHC","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 2023 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2023-06-12","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2023-11-30","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2023-12-22","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}