Bonn 2020 – scientific programme
The DPG Spring Meeting in Bonn had to be cancelled! Read more ...
Parts | Days | Selection | Search | Updates | Downloads | Help
T: Fachverband Teilchenphysik
T 62: Cosmic rays II
T 62.5: Talk
Wednesday, April 1, 2020, 17:35–17:50, L-3.002
Composition Measurements with AugerPrime using Deep Learning* — •Sonja Schröder for the Pierre Auger collaboration — Bergische Universität Wuppertal, Gaußstr. 20, 42119 Wuppertal
The AugerPrime upgrade of the Pierre Auger Observatory in Argentina enhances the precision of primary particle composition measurements made by the surface detector. This is achieved using the different responses of the Water-Cherenkov-Detector (WCD) and the Surface-Scintilator-Detector (SSD) on top, to the electromagnetic and muonic component of the extensive air shower. While the upgrade is still in progress, the cosmic ray composition sensitivity of AugerPrime can already be probed using current machine learning techniques, such as deep neural networks, on simulations.
In this presentation a deep learning approach is shown to be able to reconstruct the depth of shower maximum Xmax, a mass sensitive observable, on an event-by-event basis. A combination of deep convolutional neural networks is used to process information from both WCD and SSD signals. These signals are extracted from full AugerPrime detector simulations containing a mixed composition of protons, helium, nitrogen and iron. The sensitivity of the reconstruction will be shown, as well as its estimated bias and resolution.
* Gefördert durch die BMBF Verbundforschung Astroteilchenphysik (Vorhaben 05A17PX1).