CERN Accelerating science

Preprint
Report number	arXiv:2410.21611 ; HEPHY-ML-24-05 ; FERMILAB-PUB-24-0728-CMS ; TTK-24-43
Title	CaloChallenge 2022: A Community Challenge for Fast Calorimeter Simulation
Author(s)	Krause, Claudius (ed.) (Vienna, OAW ; Heidelberg U.) ; Faucci Giannelli, Michele (ed.) (INFN, Rome2 ; Chalmers U. Tech.) ; Kasieczka, Gregor (ed.) (Hamburg U.) ; Nachman, Benjamin (ed.) (LBNL, Berkeley) ; Salamani, Dalila (ed.) (CERN) ; Shih, David (ed.) (Rutgers U., Piscataway) ; Zaborowska, Anna (ed.) (CERN) ; Amram, Oz (Fermilab) ; Borras, Kerstin (DESY ; Aachen, Tech. Hochsch.) ; Buckley, Matthew R. (Rutgers U., Piscataway) ; Buhmann, Erik (Hamburg U.) ; Buss, Thorsten (Hamburg U. ; DESY) ; Da Costa Cardoso, Renato Paulo (CERN) ; Caterini, Anthony L. (Unlisted, CA) ; Chernyavskaya, Nadezda (CERN) ; Corchia, Federico A.G. (Bologna U. ; INFN, Bologna) ; Cresswell, Jesse C. (Unlisted, CA) ; Diefenbacher, Sascha (LBNL, Berkeley) ; Dreyer, Etienne (Weizmann Inst.) ; Ekambaram, Vijay (IBM Watson Res. Ctr.) ; Eren, Engin (DESY) ; Ernst, Florian (Heidelberg U. ; CERN) ; Favaro, Luigi (Heidelberg U.) ; Franchini, Matteo (Bologna U. ; INFN, Bologna) ; Gaede, Frank (DESY) ; Gross, Eilam (Weizmann Inst.) ; Hsu, Shih-Chieh (Washington U., Seattle) ; Jaruskova, Kristina (CERN) ; Käch, Benno (Hamburg U. ; DESY) ; Kalagnanam, Jayant (IBM Watson Res. Ctr.) ; Kansal, Raghav (Fermilab ; Caltech) ; Kim, Taewoo (Unlisted, CA) ; Kobylianskii, Dmitrii (Weizmann Inst.) ; Korol, Anatolii (DESY) ; Korcari, William (Hamburg U.) ; Krücker, Dirk (DESY) ; Krüger, Katja (DESY) ; Letizia, Marco (Genoa U. ; INFN, Genoa) ; Li, Shu (Tsung-Dao Lee Inst., Shanghai ; Shanghai Jiao Tong U.) ; Liu, Qibin (Tsung-Dao Lee Inst., Shanghai ; Shanghai Jiao Tong U.) ; Liu, Xiulong (Washington U., Seattle) ; Loaiza-Ganem, Gabriel (Unlisted, CA) ; Madula, Thandikire (University Coll. London) ; McKeown, Peter (CERN ; DESY) ; Melzer-Pellmann, Isabell-A. (DESY) ; Mikuni, Vinicius (LBNL, Berkeley) ; Nguyen, Nam (IBM Watson Res. Ctr.) ; Ore, Ayodele (Heidelberg U.) ; Palacios Schweitzer, Sofia (Heidelberg U.) ; Pang, Ian (Rutgers U., Piscataway) ; Pedro, Kevin (Fermilab) ; Plehn, Tilman (Heidelberg U.) ; Pokorski, Witold (CERN) ; Qu, Huilin (CERN) ; Raikwar, Piyush (CERN) ; Raine, John A. (Geneva U.) ; Reyes-Gonzalez, Humberto (INFN, Genoa ; Genoa U. ; RWTH Aachen U.) ; Rinaldi, Lorenzo (Bologna U. ; INFN, Bologna) ; Ross, Brendan Leigh (Unlisted, CA) ; Scham, Moritz A.W. (DESY ; Aachen, Tech. Hochsch. ; IAS, Julich) ; Schnake, Simon (DESY ; Aachen, Tech. Hochsch.) ; Shimmin, Chase (Yale U.) ; Shlizerman, Eli (Washington U., Seattle) ; Soybelman, Nathalie (Weizmann Inst.) ; Srivatsa, Mudhakar (IBM Watson Res. Ctr.) ; Tsolaki, Kalliopi (CERN) ; Vallecorsa, Sofia (CERN) ; Yeo, Kyongmin (IBM Watson Res. Ctr.) ; Zhang, Rui (Nanjing U. ; Wisconsin U., Madison)
Document contact	Contact: arXiv
Imprint	2024-10-28
Number of pages	204
Note	204 pages, 100+ figures, 30+ tables
Subject category	physics.ins-det ; Detectors and Experimental Techniques ; hep-ph ; Particle Physics - Phenomenology ; hep-ex ; Particle Physics - Experiment ; cs.LG ; Computing and Computers
Abstract	We present the results of the "Fast Calorimeter Simulation Challenge 2022" - the CaloChallenge. We study state-of-the-art generative models on four calorimeter shower datasets of increasing dimensionality, ranging from a few hundred voxels to a few tens of thousand voxels. The 31 individual submissions span a wide range of current popular generative architectures, including Variational AutoEncoders (VAEs), Generative Adversarial Networks (GANs), Normalizing Flows, Diffusion models, and models based on Conditional Flow Matching. We compare all submissions in terms of quality of generated calorimeter showers, as well as shower generation time and model size. To assess the quality we use a broad range of different metrics including differences in 1-dimensional histograms of observables, KPD/FPD scores, AUCs of binary classifiers, and the log-posterior of a multiclass classifier. The results of the CaloChallenge provide the most complete and comprehensive survey of cutting-edge approaches to calorimeter fast simulation to date. In addition, our work provides a uniquely detailed perspective on the important problem of how to evaluate generative models. As such, the results presented here should be applicable for other domains that use generative AI and require fast and faithful generation of samples in a large phase space.
Other source	Inspire
Copyright/License	preprint: (License: arXiv nonexclusive-distrib 1.0)