Speaker
Description
In recent years, radio detection of ultra-high-energy cosmic rays (UHECRs), with energies above 1018 eV, has become an established technique. Radio emissions can be simulated with high accuracy using Monte Carlo codes such as ZHAireS and CoREAS, but these simulations are computationally intensive.
In this work, we present a machine-learning-based emulator that reproduces radio signal simulations with high accuracy in milliseconds rather than hours. Using this emulator as a neural likelihood estimator, primary particle properties are inferred via simulation-based inference. On ZHAireS simulations for the GRANDProto300 array, the method achieves an 8.9% resolution on electromagnetic energy and a 0.08° angular resolution with calibrated uncertainties, matching state-of-the-art reconstruction performance.
Finally, we deploy the method on real data, successfully reconstructing cosmic-ray candidates detected by the GP300 prototype.