Fast simulation of muons produced at the SHiP experiment using Generative Adversarial Networks

Ahdida, C.; Albanese, R.; Alexandrov, A.; Anokhina, A.; Aoki, Shigeki; Arduini, G.; Atkin, E.; Azorskiy, N.; Back, J. J.; Bagulya, A.; Santos, F. Baaltasar dos; Baranov, A.; Bardou, F.; Barker, G. J.; Battistin, M.; Bauche, J.; Bay, A.; Bayliss, V.; Bencivenni, G.; Berdnikov, A.; Berdnikov, Y.; Berezkina, I.; Bertani, M.; Betancourt, C.; Bezshyiko, Ia.; Bezshyyko, Oleg; Bick, D.; Bieschke, S.; Blanco, A.; Boehm, J.; Bogomilov, M.; Bondarenko, Kyrylo; Bonivento, W.; Borburgh, J.; Boyarsky, Alexey; Brenner, R.; Bretón, D.; Brundler, R.; Bruschi, M.; Büscher, V.; Buonaura, A.; Buontempo, S.; Cadeddu, S.; Calcaterra, A.; Calviani, M.; Campanelli, M.; Casolino, M.; Charitonidis, N.; Chau, Phi; Chauveau, J.; Chepurnov, A.; Chernyavskiy, M.; Choi, Ki Young; Chumakov, A.G.; Ciambrone, P.; Congedo, L.; Cornelis, Karel; Cristinziani, M.; Crupano, A.; Dallavalle, G. M.; Datwyler, A.; D’Ambrosio, N.; D’Appollonio, Giuseppe; Saraiva, J. de Carvalho; Lellis, G. De; Magistris, M. de; Roeck, A. De; Serio, M. De; Simone, D. De; Dedenko, L. G.; Dergachev, Pavel; Crescenzo, A. Di; Marco, N. Di; Dib, Claudio O.; Dijkstra, H.; Dipinto, P.; Дмитренко, В. В.; Dmitrievskiy, S.; Dougherty, L.A.; Dolmatov, A.; Domenici, D.; Donskov, S.V.; Drohan, V.; Dubreuil, A.; Ehlert, M.; Enik, T.; Etenko, A.; Fabbri, F.; Fabbri, L.; Fabich, A.; Fedin, O. L.; Fedotovs, F.; Felici, G.; Ferro‐Luzzi, M.; Filippov, K.; Fini, R. A.; Fonte, P.; Franco, C.; Fraser, Matthew; Fresa, R.; Froeschl, R.; Fukuda, Tatsuya; Galati, G.; Gall, J.; Gatignon, L.; Gavrilov, G.; Gentile, V.; Gerlach, Stefan; Goddard, B.; Golinka-Bezshyyko, L.; Golovatiuk, A.; Golubkov, D.; Golutvin, A.; Gorbounov, P. A.; Gorbunov, D.; Горбунов, С.; Gorkavenko, Volodymyr M.; Gornushkin, Y.; Горшенков, М.В.; Grachev, Vadim; Grandchamp, A.L.; Granich, G.; Graverini, E.; Grenard, Jean-Louis; Grenier, D.; Grichine, V.; Gruzinskii, N.; Guler, A. M.; Guz, Yu.; Haefeli, G.; Hagner, C.; Hakobyan, H.; Harris, Irene; Herwijnen, E. van; Heßler, Christoph; Hollnagel, A.; Hosseini, B.; Hushchyn, M.; Iaselli, G.; Iuliano, A.; Ivantchenko, Vladimir; Jacobsson, R.; Joković, D.; Jonker, M.; Каденко, І.; Kain, V.; Kaiser, B.; Kamiscioglu, C.; Kershaw, K.; Khabibullin, M.; Khalikov, E.; Khaustov, G.; Khoriauli, G.; Khotyantsev, A.; Kim, S.H.; Kim, Y.G.; Kim, V.; Kitagawa, N.; Ko, J.-W.; Kodama, K.; Kolesnikov, A.; Kolev, D.; Kolosov, V.N.; Komatsu, M.; Kondrat’eva, N. V.; Kono, Akihiro; Konovalova, N. S.; Kormannshaus, S.; Korol, I.; Korolko, I.; Korzenev, A.; Kostyukhin, V. V.; Platia, E. Koukovini; Kovalenko, S.; Krasilnikova, I.; Kudenko, Y.; Kurbatov, E.; Kurbatov, Pavel; Kurochka, V.; Kuznetsova, E.; Lacker, H.; Lamont, Mike; Lanfranchi, G.; Lantwin, O.; Lauria, A.; Lee, K.S.; Lee, K.Y.; Lévy, J. M.; Loschiavo, V. P.; Lopes, L.; Sola, E. Lopez; Lyubovitskij, Valery E.; Maalmi, Jihane; Magnan, A.-M.; Maleev, V. P.; Малинин, А.; Manabe, Yuta; Managadze, A.K.; Manfredi, M.; Marsh, S.; Marshall, Alexander Mclean; Mefodev, A.; Mermod, P.; Miano, Andrea; Mikado, S.; Mikhaylov, Yu.; Milstead, D.; Mineev, O.; Montanari, A.; Montesi, M.C.; Morishima, Keisuke; Movchan, S.; Muttoni, Y.; Naganawa, N.; Nakamura, Masahiro; Nakano, Takashi; Nasybulin, S.; Ninin, P.; Nishio, Akira; Novikov, Alexander; Obinyakov, B.; Ogawa, S.; Okateva, N.; Opitz, B.; Osborne, John A.; Ovchynnikov, Maksym; Owtscharenko, N.; Owen, P.; Pacholek, P.; Paoloni, A.; Park, B.D.; Park, S.K.; Pastore, A.; Patel, M.; Pereyma, D.; Perillo‐Marcone, Antonio; Petkov, G.L.; Petridis, K.; Petrov, A.; Подгрудков, Д.; Poliakov, V.; Polukhina, N.; Prieto, Javier; Prokudin, M.; Prota, Andrea; Quercia, A.; Rademakers, A.; Rakai, A.; Ratnikov, F.; Rawlings, T.; Redi, F.; Ricciardi, S.; Rinaldesi, M.; Rodin, V.; Rodin, Viktor; Robbe, P.; Cavalcante, A.B. Rodrigues; Роганова, Т. М.; Rokujo, H.; Rosa, G.; Rovelli, T.; Ruchayskiy, Oleg; Ruf, T.; Samoylenko, V.D.; Samsonov, V.; Galan, F. Sanchez; Díaz, P. Santos; Ull, A. Sanz; Saputi, A.; Sato, O.; Savchenko, E. S.; Schliwinski, J.S.; Schmidt‐Parzefall, W.; Serra, N.; Sgobba, S.; Shadura, O.; Шакин, А.; Shaposhnikov, M.; Шаталов, П. Б.; Shchedrina, T. V.; Shchutska, L.; Shevchenko, V.; Shibuya, H.; Shihora, L.; Shirobokov, S.; Shustov, A.; Silverstein, S.; Simone, S.; Simoniello, R.; Skorokhvatov, M.; Smirnov, S. Yu.; Sohn, J. Y.; Sokolenko, Anastasia; Solodko, E.; Старков, Н. И.; Stoel, Linda; Storaci, B.; Stramaglia, M. E.; Sukhonos, D.; Suzuki, Y.; Takahashi, Satoshi; Tastet, Jean‐Loup; Teterin, P.; Naing, S. Than; Timiryasov, Inar; Tioukov, V.; Tommasini, D.; Torii, M.; Tosi, N.; Treille, D.; Tsenov, R.; Улин, С. Е.; Ustyuzhanin, A.; Uteshev, Z. M.; Vankova-Kirilova, G.; Vannucci, F.; Venkova, P.; Venturi, V.; Vilchinski, S.; Villa, M.; Vincke, Heinz; Vincke, Helmut; Visone, C.; Власик, К. Ф.; Volkov, A.; Voronkov, R.; Waasen, Stefan van; Wanke, R.; Wertelaers, P.; Woo, Jong-Kwan; Wurm, M.; Xella, S.; Yılmaz, Deniz; Yılmazer, A.U.; Yoon, C. S.; Zarubin, P. I.; Zarubina, I. G.; Zaytsev, Yu.

doi:10.1088/1748-0221/14/11/p11028

Cited by 40 publications

(30 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here, the authors focus on pp → 2 jet processes, and implement an Artificial Neural Network point selection (NNPS) scheme for selecting training data based on the points the network struggles to learn the most. In addition, there has been much work on the use of Generative Adversarial Networks (GANs) [14], and other generative models, for event generation [15][16][17][18][19][20][21], while there has been little work addressing the issue of explicit matrix element approximation [22].…”

Section: Introductionmentioning

confidence: 99%

Using neural networks for efficient evaluation of high multiplicity scattering amplitudes

Badger

Bullock

2020

J. High Energ. Phys.

View full text Add to dashboard Cite

Precision theoretical predictions for high multiplicity scattering rely on the evaluation of increasingly complicated scattering amplitudes which come with an extremely high CPU cost. For state-of-the-art processes this can cause technical bottlenecks in the production of fully differential distributions. In this article we explore the possibility of using neural networks to approximate multi-variable scattering amplitudes and provide efficient inputs for Monte Carlo integration. We focus on QCD corrections to e + e − → jets up to one-loop and up to five jets. We demonstrate reliable interpolation when a series of networks are trained to amplitudes that have been divided into sectors defined by their infrared singularity structure. Complete simulations for one-loop distributions show speed improvements of at least an order of magnitude over a standard approach.

show abstract

Section: Introductionmentioning

confidence: 99%

Using neural networks for efficient evaluation of high multiplicity scattering amplitudes

Badger

Bullock

2020

J. High Energ. Phys.

View full text Add to dashboard Cite

show abstract

“…Applications of GANs within particle physics are constantly appearing. GANs have been applied in both event generation [37][38][39][40][41][42][43] and detector modelling [44][45][46][47][48][49][50][51][52]. In this section the inference and training speeds of some of these particle physics based GANs are assessed on the IPU hardware and compared to results on the GPU and CPU described in Table 1.…”

Section: Event Generation and Tracking Corrections Using Gansmentioning

confidence: 99%

“…In comparing a GAN to the full simulation care needs to be taken to assign a systematic uncertainty related to the residual mismodelling. The GAN event generation is particularly helpful when the systematic uncertainty due to its mismodelling is smaller than other errors associated with other parts of the analysis procedure [38]. A limitation of the GAN-based event-generation stems from the fact that the range of the feature space that the GAN can accurately model is defined by that of the full-simulation training sample.…”

Section: Event Generationmentioning

confidence: 99%

Studying the Potential of Graphcore® IPUs for Applications in Particle Physics

Maddrell-Mander

Mohan

Marshall

et al. 2021

Comput Softw Big Sci

View full text Add to dashboard Cite

This paper presents the first study of Graphcore’s Intelligence Processing Unit (IPU) in the context of particle physics applications. The IPU is a new type of processor optimised for machine learning. Comparisons are made for neural-network-based event simulation, multiple-scattering correction, and flavour tagging, implemented on IPUs, GPUs and CPUs, using a variety of neural network architectures and hyperparameters. Additionally, a Kálmán filter for track reconstruction is implemented on IPUs and GPUs. The results indicate that IPUs hold considerable promise in addressing the rapidly increasing compute needs in particle physics.

show abstract

“…The work [16] demonstrates how optimal representation variables, describing different states of physical systems, could be learned by neural nets. Generative Adversarial Nets, a generative Deep Learning models have started to be successfully applied for the event simulations [17,18]. Review of possibilities of application of ML and DL methods to high energy physics can be found elsewhere [19][20][21].…”

Section: Jhep07(2020)133mentioning

confidence: 99%

Classification of equation of state in relativistic heavy-ion collisions using deep learning

Kvasiuk

Zabrodin

Bravina

et al. 2020

J. High Energ. Phys.

View full text Add to dashboard Cite

Convolutional Neural Nets, which is a powerful method of Deep Learning, is applied to classify equation of state of heavy-ion collision event generated within the UrQMD model. Event-by-event transverse momentum and azimuthal angle distributions of protons are used to train a classifier. An overall accuracy of classification of 98% is reached for Au+Au events at √ s N N = 11 GeV. Performance of classifiers, trained on events at different colliding energies, is investigated. Obtained results indicate extensive possibilities of application of Deep Learning methods to other problems in physics of heavyion collisions.

show abstract

Fast simulation of muons produced at the SHiP experiment using Generative Adversarial Networks

Cited by 40 publications

References 25 publications

Using neural networks for efficient evaluation of high multiplicity scattering amplitudes

Using neural networks for efficient evaluation of high multiplicity scattering amplitudes

Studying the Potential of Graphcore® IPUs for Applications in Particle Physics

Classification of equation of state in relativistic heavy-ion collisions using deep learning

Contact Info

Product

Resources

About