Cosmic-Ray Composition analysis at IceCube using Graph Neural Networks

TL;DR

This paper presents a novel graph neural network approach to analyze cosmic-ray composition using IceCube and IceTop data, enhancing the estimation of primary particle mass and understanding high-energy muon content.

Contribution

The work introduces a new graph neural network method and composition-sensitive parameters for improved cosmic-ray composition analysis at IceCube.

Findings

Enhanced accuracy in estimating cosmic-ray primary mass.

Introduction of new composition-sensitive parameters.

Potential insights into high-energy muon content in air showers.

Abstract

The IceCube Neutrino Observatory is a multi-component detector embedded deep within the South-Pole Ice. This proceeding will discuss an analysis from an integrated operation of IceCube and its surface array, IceTop, to estimate cosmic-ray composition. The work will describe a novel graph neural network based approach for estimating the mass of primary cosmic rays, that takes advantage of signal-footprint information and reconstructed cosmic-ray air shower parameters. In addition, the work will also introduce new composition-sensitive parameters for improving the estimation of cosmic-ray composition, with the potential of improving our understanding of the high-energy muon content in cosmic-ray air showers.