Toward Robustness across Cosmological Simulation Models I<scp>llustris</scp>TNG, SIMBA, A<scp>strid</scp>, and S<scp>wift</scp>-E<scp>agle</scp>

Abstract The rapid advancement of large-scale cosmological simulations has opened new avenues for cosmological and astrophysical research. However, the increasing diversity among cosmological simulation models presents a challenge to the robustness . In this work, we develop the Model-Insensitive ESTimator (M iest ), a machine that can robustly estimate the cosmological parameters, Ω m and σ 8 , from neural hydrogen maps of simulation models in the Cosmology and Astrophysics with MachinE Learning Simulations project—I llustris TNG, SIMBA , A strid , and S WIFT - E agle . An estimator is considered robust if it possesses a consistent predictive power across all simulations, including those used during the training phase. We train our machine using multiple simulation models and ensure that it only extracts common features between the models while disregarding the model-specific features. This allows us to develop a novel model that is capable of accurately estimating parameters across a range of simulation models, without being biased toward any particular model. Upon the investigation of the latent space—a set of summary statistics, we find that the implementation of robustness leads to the blending of latent variables across different models, demonstrating the removal of model-specific features. In comparison to a standard machine lacking robustness , the average performance of M iest on the unseen simulations during the training phase has been improved by ∼17% for Ω m and 38% for σ 8 . By using a machine learning approach that can extract robust , yet physical features, we hope to improve our understanding of galaxy formation and evolution in a (subgrid) model-insensitive manner, and ultimately, gain insight into the underlying physical processes responsible for robustness .