Photometric Redshift Estimation with Hybrid Quantum Neural Network

Abstract

A Hybrid Quantum Neural Network (HQNN) is applied to recover the photometric redshift of a galaxy dataset built using data from two astronomical surveys, the Kilo-Degree Survey (KIDS) and the VISTA Kilo-Degree Infrared Galaxy (VIKING), which provide data with coverage from the optical to the near-infrared electromagnetic spectrum. The results are then compared with those obtained by a classical neural network. We see that HQNN consistently recovers the reference photometric redshift with a high statistical fidelity. Furthermore, both the $R^2$ and the MSE found by the HQNN are consistently better than those found by the classical counterpart (albeit consistent in the error range), proving the worth of further investigating these novel technologies for this particular astrophysical application.