Air-Shower Reconstruction at the Pierre Auger Observatory based on Deep Learning

J. GLOMBITZA; S. J. SCIUTTO; THE PIERRE AUGER COLLABORATION.

Madison

Conferencia; 36th International Cosmic Ray Conference; 2019

The surface detector array of the Pierre Auger Observatory measures the footprint of air showersinduced by ultra-high energy cosmic rays. The reconstruction of event-by-event informationsensitive to the cosmic-ray mass, is a challenging task and so far mainly based on fluorescencedetector observations with their duty cycle of 15%. Recently, great progress has been made inmultiple fields of machine learning using deep neural networks and associated techniques. Applyingthese new techniques to air-shower physics opens up possibilities for improved reconstruction,including an estimation of the cosmic-ray composition. In this contribution, we show that deepconvolutional neural networks can be used for air-shower reconstruction, using surface-detectordata. The focus of the machine-learning algorithm is to reconstruct depths of shower maximum.In contrast to traditional reconstruction methods, the algorithm learns to extract the essential informationfrom the signal and arrival-time distributions of the secondary particles. We present theneural-network architecture, describe the training, and assess the performance using simulated airshowers.