| Preprint | |
| Report number | arXiv:2312.09290 |
| Title | Normalizing Flows for High-Dimensional Detector Simulations |
| Author(s) | Ernst, Florian (U. Heidelberg, ITP ; CERN) ; Favaro, Luigi (U. Heidelberg, ITP) ; Krause, Claudius (U. Heidelberg, ITP ; Vienna, OAW) ; Plehn, Tilman (U. Heidelberg, ITP) ; Shih, David (Rutgers U., Piscataway) |
| Imprint | 2023-12-14 |
| Number of pages | 34 |
| Note | 34 pages, 14 figures, 7 tables, journal version |
| Subject category | hep-ph ; Particle Physics - Phenomenology |
| Abstract | Whenever invertible generative networks are needed for LHC physics, normalizing flows show excellent performance. In this work, we investigate their performance for fast calorimeter shower simulations with increasing phase space dimension. We use fast and expressive coupling spline transformations applied to the CaloChallenge datasets. In addition to the base flow architecture we also employ a VAE to compress the dimensionality and train a generative network in the latent space. We evaluate our networks on several metrics, including high-level features, classifiers, and generation timing. Our findings demonstrate that invertible neural networks have competitive performance when compared to autoregressive flows, while being substantially faster during generation. |
| Other source | Inspire |
| Copyright/License | preprint: (License: CC BY 4.0) |
Element opprettet 2024-06-12, sist endret 2025-01-16
