Ultra-high energy cosmic rays are the most energetic particles found in nature and induce extensive air showers when penetrating the Earth’s atmosphere. By measuring these showers with a 3000 km² surface detector and fluorescence telescopes, the arrival directions and mass composition of the cosmic particles are studied at the Pierre Auger Observatory.
The reconstruction of event-by-event information sensitive to the cosmic-ray mass is a challenging task and so far, mainly based on fluorescence detector observations with a duty cycle of about 15%. Recently, deep learning-based algorithms have shown to be extraordinarily successful across many domains. Applying these novel algorithms to surface-detector data allows for an event-by-event estimation of the cosmic-ray mass, exploiting the 100% duty cycle of the detector.
In this contribution, we discuss the application of deep learning at the Pierre Auger Observatory with a particular focus on the reconstruction of air showers. Furthermore, we show that machine learning algorithms and the extensive capabilities of the upcoming AugerPrime upgrade have enormous potential to deepen our understanding of UHECRs.