Sub-PeV Cosmic-Ray Measurements at IceCube

The surface array of the IceCube Neutrino Observatory is able to record cosmic-ray induced air showers with a primary energy in the range from a few hundred TeV to EeV. Composition analyses at IceCube highly benefit from its multi-detector design. Combining the measurement of the electromagnetic shower component and low-energy muons at the surface with the response of the in-ice array to the associated high-energy muons improves the directional reconstruction accuracy and opens unique possibilities to extract the primary particle's mass. The analysis of air showers below PeV energies provides the potential to test the validity of hadronic interaction models at low energies but also requires special treatment in the event processing. In this talk, a new methodical approach to reconstruct the elemental composition of these low-energy air showers with a convolutional neural network is presented. The achieved performance in primary mass discrimination and energy reconstruction of air-shower events is discussed.