Machine learning based study of collective quadrupole–octupole excitations

We use the collective quadrupole–octupole model of axial symmetric shapes to study excited states in different Th and Ra isotopes. In this model, we use one of the most commonly used potentials in nuclear physics: Woods–Saxon potential. Extending this model based on the potential suggested is able to describe excited states due to quadrupole–octupole deformations satisfactorily. In this work, the necessary Hamiltonian is diagonalized in suitable bases to yield states of the candidate nuclei. This procedure also necessitates optimization, which is done using machine learning methods for each isotope. The findings and outcomes are carefully examined and contrasted with experimental values.