Constraining Interaction strength in the cross-EVHRG model through machine learning techniques

The prevalent method in the Hadron Resonance Gas (HRG) model to account for hadronic interaction is through the Van der Waals-type Excluded-Volume (EV) interaction among the hadrons. However, it's apparent that the EV-HRG approach isn't conclusive and exhibits inconsistencies even with the second-order virial expansion. To overcome this drawback, the cross HRG model is introduced, where proper interactions between every pair of particle species are taken into consideration. This enables the handling of specific two-body interactions, albeit at the expense of increased computational complexity. Various machine learning techniques are employed to tackle these challenges. In this investigation, different model parameters are fine-tuned by simultaneously considering results from lattice QCD and experimental data. The aim is to elucidate interactions across diverse hadronic sectors and understand how the intensity of these interactions correlates with the observables used to calibrate the interaction parameters.