TY  - EJOUR
AU  - Krause, Claudius
AU  - Faucci Giannelli, Michele
AU  - Kasieczka, Gregor
AU  - Nachman, Benjamin
AU  - Salamani, Dalila
AU  - Shih, David
AU  - Zaborowska, Anna
AU  - Amram, Oz
AU  - Borras, Kerstin
AU  - Buckley, Matthew R.
AU  - Buhmann, Erik
AU  - Buss, Thorsten
AU  - Da Costa Cardoso, Renato Paulo
AU  - Caterini, Anthony L.
AU  - Chernyavskaya, Nadezda
AU  - Corchia, Federico A. G.
AU  - Cresswell, Jesse C.
AU  - Diefenbacher, Sascha
AU  - Dreyer, Etienne
AU  - Ekambaram, Vijay
AU  - Eren, Engin
AU  - Ernst, Florian
AU  - Favaro, Luigi
AU  - Franchini, Matteo
AU  - Gaede, Frank
AU  - Gross, Eilam
AU  - Hsu, Shih-Chieh
AU  - Jaruskova, Kristina
AU  - Käch, Benno
AU  - Kalagnanam, Jayant
AU  - Kansal, Raghav
AU  - Kim, Taewoo
AU  - Kobylianskii, Dmitrii
AU  - Korol, Anatolii
AU  - Korcari, William
AU  - Krücker, Dirk
AU  - Krüger, Katja
AU  - Letizia, Marco
AU  - Li, Shu
AU  - Liu, Qibin
AU  - Liu, Xiulong
AU  - Loaiza-Ganem, Gabriel
AU  - Madula, Thandikire
AU  - McKeown, Peter
AU  - Melzer-Pellmann, Isabell-A.
AU  - Mikuni, Vinicius
AU  - Nguyen, Nam
AU  - Ore, Ayodele
AU  - Palacios Schweitzer, Sofia
AU  - Pang, Ian
AU  - Pedro, Kevin
AU  - Plehn, Tilman
AU  - Pokorski, Witold
AU  - Qu, Huilin
AU  - Raikwar, Piyush
AU  - Raine, John A.
AU  - Reyes-Gonzalez, Humberto
AU  - Rinaldi, Lorenzo
AU  - Ross, Brendan Leigh
AU  - Scham, Moritz A. W.
AU  - Schnake, Simon Patrik
AU  - Shimmin, Chase
AU  - Shlizerman, Eli
AU  - Soybelman, Nathalie
AU  - Srivatsa, Mudhakar
AU  - Tsolaki, Kalliopi
AU  - Vallecorsa, Sofia
AU  - Yeo, Kyongmin
AU  - Zhang, Rui
TI  - CaloChallenge 2022: A Community Challenge for Fast Calorimeter Simulation
IS  - arXiv:2410.21611
M1  - PUBDB-2025-01144
M1  - arXiv:2410.21611
M1  - HEPHY-ML-24-05
M1  - FERMILAB-PUB-24-0728-CMS
M1  - TTK-24-43
PY  - 2024
N1  - 204 pages, 100+ figures, 30+ tables
AB  - 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.
LB  - PUB:(DE-HGF)25
UR  - https://bib-pubdb1.desy.de/record/625593
ER  -