L2LFlows: generating high-fidelity 3D calorimeter images
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” (L2LFlows) - is an evolution...
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
October 18, 2023
|
| In: |
Journal of Instrumentation
Year: 2023, Jahrgang: 18, Heft: 10, Pages: 1-29 |
| ISSN: | 1748-0221 |
| DOI: | 10.1088/1748-0221/18/10/P10017 |
| Online-Zugang: | Verlag, kostenfrei, Volltext: https://doi.org/10.1088/1748-0221/18/10/P10017 Verlag, kostenfrei, Volltext: https://dx.doi.org/10.1088/1748-0221/18/10/P10017 
|
| Verfasserangaben: | Sascha Diefenbacher, Engin Eren, Frank Gaede, Gregor Kasieczka, Claudius Krause, Imahn Shekhzadeh and David Shih |
| Zusammenfassung: | 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” (L2LFlows) - is an evolution of the CaloFlow architecture adapted to a higher-dimensional setting (30 layers of 10× 10 voxels each). The main innovation of L2LFlows 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. |
|---|---|
| Beschreibung: | Gesehen am 31.07.2024 |
| Beschreibung: | Online Resource |
| ISSN: | 1748-0221 |
| DOI: | 10.1088/1748-0221/18/10/P10017 |