Modern ML models trained on simulations often degrade on real data because of domain shift. I will present a semi-supervised domain adaptation (SSDA) pipeline that transfers a four-class pseudo-spectral classifier (high-z QSOs, low-z QSOs, galaxies, stars) from abundant DESI→J-PAS mocks (~1.5M) to real J-PAS observations using only a small labeled J-PAS subset. The method pretrains on mocks,...
Understanding the formation and evolution of galaxies over cosmic time requires a
comprehensive analysis of their morphologies, especially because morphological
features are strongly connected to other galaxy properties such as stellar
populations, environments, and kinematics. However, the growing size of modern
sky surveys has resulted in massive volumes of unclassified galaxies,...
We present a study on the application of different machine learning algorithms for the identification of hypernuclei produced in heavy-ion collisions, particularly those with mass numbers A = 3 to A = 5. The study focuses on three supervised learning algorithms - Boosted Decision Trees, Support Vector Machines, and Artificial Neural Networks - which were trained to distinguish true hypernuclei...
Gravitational wave sources with electromagnetic counterparts have highlighted the need for predictive, interpretable models linking the parameters of compact binary systems to post-merger remnants and mass outflows. In this work, we explore AI-driven symbolic regression (SR) frameworks to derive updated analytical relations for disk ejecta mass in binary neutron star mergers, trained on...
