L2LFlows: Generating High-Fidelity 3D Calorimeter Images
| Autor: | Diefenbacher, Sascha, Eren, Engin, Gaede, Frank, Kasieczka, Gregor, Krause, Claudius, Shekhzadeh, Imahn, Shih, David |
|---|---|
| Rok vydání: | 2023 |
| Předmět: | |
| Zdroj: | 2023 JINST 18 P10017 |
| Druh dokumentu: | Working Paper |
| DOI: | 10.1088/1748-0221/18/10/P10017 |
| Popis: | We explore the use of normalizing flows to emulate Monte Carlo detector simulations of photon showers in a high-granularity electromagnetic calorimeter prototype for the International Large Detector (ILD). Our proposed method -- which we refer to as "Layer-to-Layer-Flows" (L$2$LFlows) -- is an evolution of the CaloFlow architecture adapted to a higher-dimensional setting (30 layers of $10\times 10$ voxels each). The main innovation of L$2$LFlows consists of introducing $30$ separate normalizing flows, one for each layer of the calorimeter, where each flow is conditioned on the previous five layers in order to learn the layer-to-layer correlations. We compare our results to the BIB-AE, a state-of-the-art generative network trained on the same dataset and find our model has a significantly improved fidelity. Comment: v2: 28 pages, 13 figures; matches version accepted for publication in JINST. Neither SISSA Medialab Srl nor IOP Publishing Ltd is responsible for any errors or omissions in this version of the manuscript or any version derived from it. Published version available via DOI |
| Databáze: | arXiv |
| Externí odkaz: |
|