CaloChallenge 2022: A Community Challenge for Fast Calorimeter Simulation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909899934400512 |
|---|---|
| author | Krause, Claudius Giannelli, Michele Faucci Kasieczka, Gregor Nachman, Benjamin Salamani, Dalila Shih, David Zaborowska, Anna Amram, Oz Borras, Kerstin Buckley, Matthew R. Buhmann, Erik Buss, Thorsten Cardoso, Renato Paulo Da Costa Caterini, Anthony L. Chernyavskaya, Nadezda Corchia, Federico A. G. Cresswell, Jesse C. Diefenbacher, Sascha Dreyer, Etienne Ekambaram, Vijay Eren, Engin Ernst, Florian Favaro, Luigi Franchini, Matteo Gaede, Frank Gross, Eilam Hsu, Shih-Chieh Jaruskova, Kristina Käch, Benno Kalagnanam, Jayant Kansal, Raghav Kim, Taewoo Kobylianskii, Dmitrii Korol, Anatolii Korcari, William Krücker, Dirk Krüger, Katja Letizia, Marco Li, Shu Liu, Qibin Liu, Xiulong Loaiza-Ganem, Gabriel Madula, Thandikire McKeown, Peter Melzer-Pellmann, Isabell-A. Mikuni, Vinicius Nguyen, Nam Ore, Ayodele Schweitzer, Sofia Palacios Pang, Ian Pedro, Kevin Plehn, Tilman Pokorski, Witold Qu, Huilin Raikwar, Piyush Raine, John A. Reyes-Gonzalez, Humberto Rinaldi, Lorenzo Ross, Brendan Leigh Scham, Moritz A. W. Schnake, Simon Shimmin, Chase Shlizerman, Eli Soybelman, Nathalie Srivatsa, Mudhakar Tsolaki, Kalliopi Vallecorsa, Sofia Yeo, Kyongmin Zhang, Rui |
| author_facet | Krause, Claudius Giannelli, Michele Faucci Kasieczka, Gregor Nachman, Benjamin Salamani, Dalila Shih, David Zaborowska, Anna Amram, Oz Borras, Kerstin Buckley, Matthew R. Buhmann, Erik Buss, Thorsten Cardoso, Renato Paulo Da Costa Caterini, Anthony L. Chernyavskaya, Nadezda Corchia, Federico A. G. Cresswell, Jesse C. Diefenbacher, Sascha Dreyer, Etienne Ekambaram, Vijay Eren, Engin Ernst, Florian Favaro, Luigi Franchini, Matteo Gaede, Frank Gross, Eilam Hsu, Shih-Chieh Jaruskova, Kristina Käch, Benno Kalagnanam, Jayant Kansal, Raghav Kim, Taewoo Kobylianskii, Dmitrii Korol, Anatolii Korcari, William Krücker, Dirk Krüger, Katja Letizia, Marco Li, Shu Liu, Qibin Liu, Xiulong Loaiza-Ganem, Gabriel Madula, Thandikire McKeown, Peter Melzer-Pellmann, Isabell-A. Mikuni, Vinicius Nguyen, Nam Ore, Ayodele Schweitzer, Sofia Palacios Pang, Ian Pedro, Kevin Plehn, Tilman Pokorski, Witold Qu, Huilin Raikwar, Piyush Raine, John A. Reyes-Gonzalez, Humberto Rinaldi, Lorenzo Ross, Brendan Leigh Scham, Moritz A. W. Schnake, Simon Shimmin, Chase Shlizerman, Eli Soybelman, Nathalie Srivatsa, Mudhakar Tsolaki, Kalliopi Vallecorsa, Sofia Yeo, Kyongmin Zhang, Rui |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_21611 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CaloChallenge 2022: A Community Challenge for Fast Calorimeter Simulation Krause, Claudius Giannelli, Michele Faucci Kasieczka, Gregor Nachman, Benjamin Salamani, Dalila Shih, David Zaborowska, Anna Amram, Oz Borras, Kerstin Buckley, Matthew R. Buhmann, Erik Buss, Thorsten Cardoso, Renato Paulo Da Costa Caterini, Anthony L. Chernyavskaya, Nadezda Corchia, Federico A. G. Cresswell, Jesse C. Diefenbacher, Sascha Dreyer, Etienne Ekambaram, Vijay Eren, Engin Ernst, Florian Favaro, Luigi Franchini, Matteo Gaede, Frank Gross, Eilam Hsu, Shih-Chieh Jaruskova, Kristina Käch, Benno Kalagnanam, Jayant Kansal, Raghav Kim, Taewoo Kobylianskii, Dmitrii Korol, Anatolii Korcari, William Krücker, Dirk Krüger, Katja Letizia, Marco Li, Shu Liu, Qibin Liu, Xiulong Loaiza-Ganem, Gabriel Madula, Thandikire McKeown, Peter Melzer-Pellmann, Isabell-A. Mikuni, Vinicius Nguyen, Nam Ore, Ayodele Schweitzer, Sofia Palacios Pang, Ian Pedro, Kevin Plehn, Tilman Pokorski, Witold Qu, Huilin Raikwar, Piyush Raine, John A. Reyes-Gonzalez, Humberto Rinaldi, Lorenzo Ross, Brendan Leigh Scham, Moritz A. W. Schnake, Simon Shimmin, Chase Shlizerman, Eli Soybelman, Nathalie Srivatsa, Mudhakar Tsolaki, Kalliopi Vallecorsa, Sofia Yeo, Kyongmin Zhang, Rui Instrumentation and Detectors Machine Learning High Energy Physics - Experiment High Energy Physics - Phenomenology 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. |
| title | CaloChallenge 2022: A Community Challenge for Fast Calorimeter Simulation |
| topic | Instrumentation and Detectors Machine Learning High Energy Physics - Experiment High Energy Physics - Phenomenology |
| url | https://arxiv.org/abs/2410.21611 |