Deep Generative Models for Fast Photon Shower Simulation in ATLAS

The ATLAS collaboration

doi:10.1007/s41781-023-00106-9

Deep Generative Models for Fast Photon Shower Simulation in ATLAS

The ATLAS collaboration

Radiation Imaging and Advanced Diagnosti

Research output: Contribution to journal › Article › peer-review

12 Scopus citations

Abstract

The need for large-scale production of highly accurate simulated event samples for the extensive physics programme of the ATLAS experiment at the Large Hadron Collider motivates the development of new simulation techniques. Building on the recent success of deep learning algorithms, variational autoencoders and generative adversarial networks are investigated for modelling the response of the central region of the ATLAS electromagnetic calorimeter to photons of various energies. The properties of synthesised showers are compared with showers from a full detector simulation using geant4. Both variational autoencoders and generative adversarial networks are capable of quickly simulating electromagnetic showers with correct total energies and stochasticity, though the modelling of some shower shape distributions requires more refinement. This feasibility study demonstrates the potential of using such algorithms for ATLAS fast calorimeter simulation in the future and shows a possible way to complement current simulation techniques.

Original language	English
Article number	7
Journal	Computing and Software for Big Science
Volume	8
Issue number	1
DOIs	https://doi.org/10.1007/s41781-023-00106-9
State	Published - Dec 2024

Funding

We acknowledge the support of ANPCyT, Argentina; YerPhI, Armenia; ARC, Australia; BMWFW and FWF, Austria; ANAS, Azerbaijan; CNPq and FAPESP, Brazil; NSERC, NRC and CFI, Canada; CERN; ANID, Chile; CAS, MOST and NSFC, China; Minciencias, Colombia; MEYS CR, Czech Republic; DNRF and DNSRC, Denmark; IN2P3-CNRS and CEA-DRF/IRFU, France; SRNSFG, Georgia; BMBF, HGF and MPG, Germany; GSRI, Greece; RGC and Hong Kong SAR, China; ISF and Benoziyo Center, Israel; INFN, Italy; MEXT and JSPS, Japan; CNRST, Morocco; NWO, Netherlands; RCN, Norway; MEiN, Poland; FCT, Portugal; MNE/IFA, Romania; MESTD, Serbia; MSSR, Slovakia; ARRS and MIZ\u0160, Slovenia; DSI/NRF, South Africa; MICINN, Spain; SRC and Wallenberg Foundation, Sweden; SERI, SNSF and Cantons of Bern and Geneva, Switzerland; MOST, Taiwan; TENMAK, T\u00FCrkiye; STFC, United Kingdom; DOE and NSF, United States of America. In addition, individual groups and members have received support from BCKDF, CANARIE, Compute Canada and CRC, Canada; PRIMUS 21/SCI/017 and UNCE SCI/013, Czech Republic; COST, ERC, ERDF, Horizon 2020 and Marie Sk\u0142odowska-Curie Actions, European Union; Investissements d\u2019Avenir Labex, Investissements d\u2019Avenir Idex and ANR, France; DFG and AvH Foundation, Germany; Herakleitos, Thales and Aristeia programmes co-financed by EU-ESF and the Greek NSRF, Greece; BSF-NSF and MINERVA, Israel; Norwegian Financial Mechanism 2014-2021, Norway; NCN and NAWA, Poland; La Caixa Banking Foundation, CERCA Programme Generalitat de Catalunya and PROMETEO and GenT Programmes Generalitat Valenciana, Spain; G\u00F6ran Gustafssons Stiftelse, Sweden; The Royal Society and Leverhulme Trust, United Kingdom. We thank CERN for the very successful operation of the LHC, as well as the support staff from our institutions without whom ATLAS could not be operated efficiently. We acknowledge the support of ANPCyT, Argentina; YerPhI, Armenia; ARC, Australia; BMWFW and FWF, Austria; ANAS, Azerbaijan; CNPq and FAPESP, Brazil; NSERC, NRC and CFI, Canada; CERN; ANID, Chile; CAS, MOST and NSFC, China; Minciencias, Colombia; MEYS CR, Czech Republic; DNRF and DNSRC, Denmark; IN2P3-CNRS and CEA-DRF/IRFU, France; SRNSFG, Georgia; BMBF, HGF and MPG, Germany; GSRI, Greece; RGC and Hong Kong SAR, China; ISF and Benoziyo Center, Israel; INFN, Italy; MEXT and JSPS, Japan; CNRST, Morocco; NWO, Netherlands; RCN, Norway; MEiN, Poland; FCT, Portugal; MNE/IFA, Romania; MESTD, Serbia; MSSR, Slovakia; ARRS and MIZ\u0160, Slovenia; DSI/NRF, South Africa; MICINN, Spain; SRC and Wallenberg Foundation, Sweden; SERI, SNSF and Cantons of Bern and Geneva, Switzerland; MOST, Taiwan; TENMAK, T\u00FCrkiye; STFC, United Kingdom; DOE and NSF, United States of America. In addition, individual groups and members have received support from BCKDF, CANARIE, Compute Canada and CRC, Canada; PRIMUS 21/SCI/017 and UNCE SCI/013, Czech Republic; COST, ERC, ERDF, Horizon 2020 and Marie Sk\u0142odowska-Curie Actions, European Union; Investissements d\u2019Avenir Labex, Investissements d\u2019Avenir Idex and ANR, France; DFG and AvH Foundation, Germany; Herakleitos, Thales and Aristeia programmes co-financed by EU-ESF and the Greek NSRF, Greece; BSF-NSF and MINERVA, Israel; Norwegian Financial Mechanism 2014-2021, Norway; NCN and NAWA, Poland; La Caixa Banking Foundation, CERCA Programme Generalitat de Catalunya and PROMETEO and GenT Programmes Generalitat Valenciana, Spain; G\u00F6ran Gustafssons Stiftelse, Sweden; The Royal Society and Leverhulme Trust, United Kingdom. The crucial computing support from all WLCG partners is acknowledged gratefully, in particular from CERN, the ATLAS Tier-1 facilities at TRIUMF (Canada), NDGF (Denmark, Norway, Sweden), CC-IN2P3 (France), KIT/GridKA (Germany), INFN-CNAF (Italy), NL-T1 (Netherlands), PIC (Spain), ASGC (Taiwan), RAL (UK) and BNL (USA), the Tier-2 facilities worldwide and large non-WLCG resource providers. Major contributors of computing resources are listed in Ref.\u00A0[51].

Access to Document

10.1007/s41781-023-00106-9

Cite this

@article{b3d27a6c76f64c5db8246beead73020e,

title = "Deep Generative Models for Fast Photon Shower Simulation in ATLAS",

abstract = "The need for large-scale production of highly accurate simulated event samples for the extensive physics programme of the ATLAS experiment at the Large Hadron Collider motivates the development of new simulation techniques. Building on the recent success of deep learning algorithms, variational autoencoders and generative adversarial networks are investigated for modelling the response of the central region of the ATLAS electromagnetic calorimeter to photons of various energies. The properties of synthesised showers are compared with showers from a full detector simulation using geant4. Both variational autoencoders and generative adversarial networks are capable of quickly simulating electromagnetic showers with correct total energies and stochasticity, though the modelling of some shower shape distributions requires more refinement. This feasibility study demonstrates the potential of using such algorithms for ATLAS fast calorimeter simulation in the future and shows a possible way to complement current simulation techniques.",

author = "{The ATLAS collaboration} and G. Aad and B. Abbott and Abbott, {D. C.} and Abud, {A. Abed} and K. Abeling and Abhayasinghe, {D. K.} and Abidi, {S. H.} and A. Aboulhorma and H. Abramowicz and H. Abreu and Y. Abulaiti and Hoffman, {A. C.Abusleme} and Acharya, {B. S.} and B. Achkar and L. Adam and Bourdarios, {C. Adam} and L. Adamczyk and L. Adamek and Addepalli, {S. V.} and J. Adelman and A. Adiguzel and S. Adorni and T. Adye and Affolder, {A. A.} and Y. Afik and Agaras, {M. N.} and J. Agarwala and A. Aggarwal and C. Agheorghiesei and Aguilar-Saavedra, {J. A.} and A. Ahmad and F. Ahmadov and Ahmed, {W. S.} and X. Ai and G. Aielli and I. Aizenberg and M. Akbiyik and {\AA}kesson, {T. P.A.} and Akimov, {A. V.} and Khoury, {K. Al} and Alberghi, {G. L.} and J. Albert and P. Albicocco and Verzini, {M. J.Alconada} and S. Alderweireldt and M. Aleksa and Aleksandrov, {I. N.} and C. Alexa and T. Alexopoulos and Steinhebel, {A. L.}",

note = "Publisher Copyright: {\textcopyright} The Author(s) 2024.",

year = "2024",

month = dec,

doi = "10.1007/s41781-023-00106-9",

language = "English",

volume = "8",

journal = "Computing and Software for Big Science",

issn = "2510-2044",

publisher = "Springer",

number = "1",

}

TY - JOUR

T1 - Deep Generative Models for Fast Photon Shower Simulation in ATLAS

AU - The ATLAS collaboration

AU - Aad, G.

AU - Abbott, B.

AU - Abbott, D. C.

AU - Abud, A. Abed

AU - Abeling, K.

AU - Abhayasinghe, D. K.

AU - Abidi, S. H.

AU - Aboulhorma, A.

AU - Abramowicz, H.

AU - Abreu, H.

AU - Abulaiti, Y.

AU - Hoffman, A. C.Abusleme

AU - Acharya, B. S.

AU - Achkar, B.

AU - Adam, L.

AU - Bourdarios, C. Adam

AU - Adamczyk, L.

AU - Adamek, L.

AU - Addepalli, S. V.

AU - Adelman, J.

AU - Adiguzel, A.

AU - Adorni, S.

AU - Adye, T.

AU - Affolder, A. A.

AU - Afik, Y.

AU - Agaras, M. N.

AU - Agarwala, J.

AU - Aggarwal, A.

AU - Agheorghiesei, C.

AU - Aguilar-Saavedra, J. A.

AU - Ahmad, A.

AU - Ahmadov, F.

AU - Ahmed, W. S.

AU - Ai, X.

AU - Aielli, G.

AU - Aizenberg, I.

AU - Akbiyik, M.

AU - Åkesson, T. P.A.

AU - Akimov, A. V.

AU - Khoury, K. Al

AU - Alberghi, G. L.

AU - Albert, J.

AU - Albicocco, P.

AU - Verzini, M. J.Alconada

AU - Alderweireldt, S.

AU - Aleksa, M.

AU - Aleksandrov, I. N.

AU - Alexa, C.

AU - Alexopoulos, T.

AU - Steinhebel, A. L.

PY - 2024/12

Y1 - 2024/12

N2 - The need for large-scale production of highly accurate simulated event samples for the extensive physics programme of the ATLAS experiment at the Large Hadron Collider motivates the development of new simulation techniques. Building on the recent success of deep learning algorithms, variational autoencoders and generative adversarial networks are investigated for modelling the response of the central region of the ATLAS electromagnetic calorimeter to photons of various energies. The properties of synthesised showers are compared with showers from a full detector simulation using geant4. Both variational autoencoders and generative adversarial networks are capable of quickly simulating electromagnetic showers with correct total energies and stochasticity, though the modelling of some shower shape distributions requires more refinement. This feasibility study demonstrates the potential of using such algorithms for ATLAS fast calorimeter simulation in the future and shows a possible way to complement current simulation techniques.

AB - The need for large-scale production of highly accurate simulated event samples for the extensive physics programme of the ATLAS experiment at the Large Hadron Collider motivates the development of new simulation techniques. Building on the recent success of deep learning algorithms, variational autoencoders and generative adversarial networks are investigated for modelling the response of the central region of the ATLAS electromagnetic calorimeter to photons of various energies. The properties of synthesised showers are compared with showers from a full detector simulation using geant4. Both variational autoencoders and generative adversarial networks are capable of quickly simulating electromagnetic showers with correct total energies and stochasticity, though the modelling of some shower shape distributions requires more refinement. This feasibility study demonstrates the potential of using such algorithms for ATLAS fast calorimeter simulation in the future and shows a possible way to complement current simulation techniques.

UR - http://www.scopus.com/inward/record.url?scp=85189330049&partnerID=8YFLogxK

U2 - 10.1007/s41781-023-00106-9

DO - 10.1007/s41781-023-00106-9

M3 - Article

AN - SCOPUS:85189330049

SN - 2510-2044

VL - 8

JO - Computing and Software for Big Science

JF - Computing and Software for Big Science

IS - 1

M1 - 7

ER -

Deep Generative Models for Fast Photon Shower Simulation in ATLAS

Abstract

Funding

Access to Document

Other files and links

Fingerprint

Cite this