2025A&A...701A.162T

Context. Modern sky surveys produce vast amounts of observational data, which makes the application of classical methods for estimating galaxy properties challenging and time-consuming. This challenge can be significantly alleviated by employing automatic machine- and deep-learning techniques.
Aims. We propose an implementation of the ULISSE algorithm to determine the physical parameters of galaxies, in particular, the star formation rates (SFR) and stellar masses (ℳ_*), based on composite-colour images alone.
Methods.ULISSE is able to rapidly and efficiently identify candidates from a single image based on photometric and morphological similarities to a given reference object with known properties. This approach leverages features extracted from the ImageNet dataset to perform similarity searches among all objects in the sample. This eliminates the need for extensive neural-network training.
Results. Our experiments were performed on the Sloan Digital Sky Survey. They demonstrate that we are able to predict the joint SFR and ℳ_* of the target galaxies within 1 dex in 60% to 80% of cases, depending on the investigated subsample (quiescent and starforming galaxies, early- and late-type, etc.), and within 0.5 dex when we consider these parameters separately. This is approximately twice the fraction obtained from a random guess extracted from the parent population. Additionally, we found that ULISSE is more effective for galaxies with an active star formation than for elliptical galaxies with quenched star formation. Additionally, ULISSE performs more efficiently for galaxies with bright nuclei such as active galactic nuclei.
Conclusions. Our results suggest that ULISSE is a promising tool for a preliminary estimation of SFR and ℳ_* for galaxies based only on single images in current and future wide-field surveys (e.g. Euclid and LSST), which target millions of sources nightly.