Machine Learning Applications at the Pierre Auger Observatory

The Pierre Auger Observatory is utilised to study the extensive air showers produced by ultra-high-energy cosmic rays. In this contribution, we provide an overview of the machine-learning techniques used by the Pierre Auger Collaboration to improve the understanding of data measured by the surface detector of the Observatory. The three methods presented use the spatial and temporal information contained in the signals measured by the surface detector stations. The first method demonstrates the application of deep learning techniques to reconstruct the energy of the cosmic ray. It has the potential to improve upon the standard technique by reducing the dependency of the energy on the primary mass. One of the primary objectives of the Observatory is to understand the evolution of the mass composition with energy. We can achieve this using observables such as the depth of the shower maximum and the number of muons reaching the ground. In the second method presented, long short-term memory and convolutional neural networks are employed to determine the depth of the shower maximum. The third work focuses on estimating both the depth of the shower maximum and the number of muons combining signals from the upgraded stations of the surface detector. Transformer networks are used for this purpose. Using simulations, we study the potential for an accurate reconstruction of the primary mass by combining the measurements of the shower maximum and the muon number.