Boosting the perf. of NNs using symmetry props. for the prediction of the shower maximum using the SD of the Pierre Auger Obs.

Discussion timeslot (ZOOM-Meeting): 16. July 2021 - 18:00
ZOOM-Meeting URL: https://icrc2021.desy.de/pf_access_abstracts
Corresponding Session: https://icrc2021-venue.desy.de/channel/Presenter-Forum-1-Evening-All-Categories/48
Abstract:
'To probe physics beyond the scales of human-made accelerators with cosmic rays demands an accurate knowledge of their primary mass composition. Using fluorescence detectors, one is able to estimate this by measuring the depth of the shower maximum $X_text{max}$. These, however, exhibit a very low duty cycle of typically below 15 %. Inferring $X_text{max}$ from a surface detector array (SD) such as the water-Cherenkov array of the Pierre Auger Observatory is highly non-trivial due to the inherent complexity and fluctuations of the shower footprint. Moreover, the sheer amount of data makes it non-trivial to find hidden patterns in the spatial and temporal distributions of detector signals. Neural networks provide a straightforward way of tackling such a problem doing a data-driven analysis. rnrnRelying solely on geometrical quantities, timing, and the signal-time information of the SD stations, we show that by exploiting the symmetries due to their triangular arrangement, we are able to boost a standard analysis network significantly without modifying its architecture or training process. Furthermore, these considerations yield a standardization procedure which also enables us to encode the footprint’s information in a memory-efficient way. The presented procedure can also be generalized and extended to systems whose setup has an underlying hexagonal geometry.'

Authors: Steffen Hahn | Markus Roth | David Schmidt | Darko Veberič
Indico-ID: 834
Proceeding URL: https://pos.sissa.it/395/239