Denoising Medium Resolution Stellar Spectra with U-Net Convolutional Neural Networks

We investigated the use of a U-Net convolutional neural network for denoising simulated medium-resolution stellar spectroscopic observations, generated under realistic observational conditions resembling the Subaru Prime Focus Spectrograph (PFS). We found that our U-Net model effectively captured spectral features, achieving an average relative error of around \(1\%\) across diverse stellar parameters, despite a limited training set of only \(1000\) observations and a relatively short training period. Although U-Net did not reach the performance previously demonstrated by denoising autoencoders (DAEs) trained extensively on larger datasets, it outperformed DAEs trained under similarly constrained conditions. These results indicate that the U-Net architecture offers rapid, robust feature learning and may be particularly advantageous in scenarios involving initial denoising, subsequently refined by more accurate, but otherwise slower deep-learning models. Abstract Published by the Jagiellonian University 2025 authors