Inverse design with deep generative models: next step in materials discovery

In the past few years, machine-learning(ML) techniques have been extensively applied in material discovery. Such techniques are applied to minimize the computationally or experimentally expensive costs of the research process, greatly reducing overall design time [1]. Typically,ML algorithms are combined with traditional methods like first-principle calculations to accelerate the optimization of compositions on the known crystal structures(elemental substitution) in the database or literature, or to search for new configurations of a fixed chemical composition.