Neural networks are known to be universal approximants for any function in an arbitrary number of variables. This property has been exploited in recent years in conjunction with Variational Monte Carlo methods and efficient optimization techniques to obtain excellent representations of the ground state of many-body quantum systems. Such wavefunctions are known as neural quantum states (NQS). Application of NQS to the study of medium-light nuclei provided very accurate energies and estimates of other observables. Among them it is possible to compute matrix elements to be used in the study of electroweak processes, as for instance beta decay, both in the hadronic and in the leptonic sectors. The Ph.D. student will become an expert in the use of such techniques, providing results that will be exploited in the analysis of experimental data and in view of applications in astrophysical contexts and in the search of physics beyond the Standard Model.
