CERN Accelerating science

Article
Report number	arXiv:2105.14027 ; FERMILAB-PUB-21-285-CMS
Title	The Dark Machines Anomaly Score Challenge: Benchmark Data and Model Independent Event Classification for the Large Hadron Collider
Author(s)	Aarrestad, Thea (CERN) ; van Beekveld, Melissa (Oxford U., Theor. Phys.) ; Bona, Marcella (Queen Mary, U. of London) ; Boveia, Antonio (Ohio State U.) ; Caron, Sascha (Nikhef, Amsterdam) ; Davies, Joe (Queen Mary, U. of London) ; de Simone, Andrea (SISSA, Trieste ; INFN, Trieste) ; Doglioni, Caterina (Lund U.) ; Duarte, Javier (UC, San Diego) ; Farbin, Amir (Texas U., Arlington) ; Gupta, Honey (Google Inc.) ; Hendriks, Luc (Nikhef, Amsterdam) ; Heinrich, Lukas A. (CERN) ; Howarth, James (Glasgow U.) ; Jawahar, Pratik (Worcester Poly. ; CERN) ; Jueid, Adil (Konkuk U.) ; Lastow, Jessica (Lund U.) ; Leinweber, Adam (Adelaide U.) ; Mamuzic, Judita (Valencia U., IFIC) ; Merényi, Erzsébet (Rice U.) ; Morandini, Alessandro (RWTH Aachen U.) ; Moskvitina, Polina (Nikhef, Amsterdam) ; Nellist, Clara (Nikhef, Amsterdam) ; Ngadiuba, Jennifer (Fermilab ; Caltech) ; Ostdiek, Bryan (Harvard U. ; IAIFI, Cambridge) ; Pierini, Maurizio (CERN) ; Ravina, Baptiste (Glasgow U.) ; de Austri, Roberto Ruiz (Valencia U., IFIC) ; Sekmen, Sezen (Kyungpook Natl. U.) ; Touranakou, Mary (Athens U. ; CERN) ; Vaškeviciute, Marija (Glasgow U.) ; Vilalta, Ricardo (Houston U.) ; Vlimant, Jean-Roch (Caltech) ; Verheyen, Rob (University Coll. London) ; White, Martin (Adelaide U.) ; Wulff, Eric (Lund U.) ; Wallin, Erik (Lund U.) ; Wozniak, Kinga A. (Vienna U. ; CERN) ; Zhang, Zhongyi (Nikhef, Amsterdam) 顯示全部 39 名作者
Publication	2022-01-28
Imprint	2021-05-28
Number of pages	57
Note	v1: 54 pages, 24 figures. v2: 56 pages, citations added, extend discussion of look-elsewhere-effect, results unchanged; v3. minor typos and updated references
In:	SciPost Phys. 12 (2022) 043
DOI	10.21468/SciPostPhys.12.1.043 (publication)
Subject category	stat.ML ; Mathematical Physics and Mathematics ; physics.data-an ; Other Fields of Physics ; hep-ex ; Particle Physics - Experiment ; hep-ph ; Particle Physics - Phenomenology
Abstract	We describe the outcome of a data challenge conducted as part of the Dark Machines Initiative and the Les Houches 2019 workshop on Physics at TeV colliders. The challenged aims at detecting signals of new physics at the LHC using unsupervised machine learning algorithms. First, we propose how an anomaly score could be implemented to define model-independent signal regions in LHC searches. We define and describe a large benchmark dataset, consisting of >1 Billion simulated LHC events corresponding to $10 \rm{fb}^{-1}$ of proton-proton collisions at a center-of-mass energy of 13 TeV. We then review a wide range of anomaly detection and density estimation algorithms, developed in the context of the data challenge, and we measure their performance in a set of realistic analysis environments. We draw a number of useful conclusions that will aid the development of unsupervised new physics searches during the third run of the LHC, and provide our benchmark dataset for future studies at https://www.phenoMLdata.org. Code to reproduce the analysis is provided at https://github.com/bostdiek/DarkMachines-UnsupervisedChallenge.
Copyright/License	preprint: (License: CC BY 4.0) publication: © Authors

Corresponding record in: Inspire