학술논문
Unbinned Deep Learning Jet Substructure Measurement in High $Q^2$ ep collisions at HERA
Document Type
article
Author
The H1 collaboration; Andreev, V; Arratia, M; Baghdasaryan, A; Baty, A; Begzsuren, K; Bolz, A; Boudry, V; Brandt, G; Britzger, D; Buniatyan, A; Bystritskaya, L; Campbell, AJ; Avila, KB Cantun; Cerny, K; Chekelian, V; Chen, Z; Contreras, JG; Cvach, J; Dainton, JB; Daum, K; Deshpande, A; Diaconu, C; Drees, A; Eckerlin, G; Egli, S; Elsen, E; Favart, L; Fedotov, A; Feltesse, J; Fleischer, M; Fomenko, A; Gal, C; Gayler, J; Goerlich, L; Gogitidze, N; Gouzevitch, M; Grab, C; Greenshaw, T; Grindhammer, G; Haidt, D; Henderson, RCW; Hessler, J; Hladký, J; Hoffmann, D; Horisberger, R; Hreus, T; Huber, F; Jacobs, PM; Jacquet, M; Janssen, T; Jung, AW; Katzy, J; Kiesling, C; Klein, M; Kleinwort, C; Klest, HT; Kogler, R; Kostka, P; Kretzschmar, J; Krücker, D; Krüger, K; Landon, MPJ; Lange, W; Laycock, P; Lee, SH; Levonian, S; Li, W; Lin, J; Lipka, K; List, B; List, J; Lobodzinski, B; Long, OR; Malinovski, E; Martyn, H-U; Maxfield, SJ; Mehta, A; Meyer, AB; Meyer, J; Mikocki, S; Mikuni, VM; Mondal, MM; Müller, K; Nachman, B; Naumann, Th; Newman, PR; Niebuhr, C; Nowak, G; Olsson, JE; Ozerov, D; Park, S; Pascaud, C; Patel, GD; Perez, E; Petrukhin, A; Picuric, I; Pitzl, D; Polifka, R; Preins, S
Source
Subject
Language
Abstract
The radiation pattern within high energy quark- and gluon-initiated jets (jetsubstructure) is used extensively as a precision probe of the strong force aswell as an environment for optimizing event generators with numerousapplications in high energy particle and nuclear physics. Looking atelectron-proton collisions is of particular interest as many of thecomplications present at hadron colliders are absent. A detailed study ofmodern jet substructure observables, jet angularities, in electron-protoncollisions is presented using data recorded using the H1 detector at HERA. Themeasurement is unbinned and multi-dimensional, using machine learning tocorrect for detector effects. All of the available reconstructed objectinformation of the respective jets is interpreted by a graph neural network,achieving superior precision on a selected set of jet angularities. Trainingthese networks was enabled by the use of a large number of GPUs in thePerlmutter supercomputer at Berkeley Lab. The particle jets are reconstructedin the laboratory frame, using the $k_{\mathrm{T}}$ jet clustering algorithm.Results are reported at high transverse momentum transfer $Q^2>150$ GeV${}^2$,and inelasticity $0.2 < y < 0.7$. The analysis is also performed in sub-regionsof $Q^2$, thus probing scale dependencies of the substructure variables. Thedata are compared with a variety of predictions and point towards possibleimprovements of such models.