Cluster Generation via Deep Energy-Based Model
This provides a universal method for materials science researchers to predict stable nanocluster configurations, including larger sizes not seen in training, though it builds incrementally on deep learning approaches.
The paper tackles the problem of generating stable nanocluster structures by constructing a smooth artificial potential energy surface using graph convolutional networks, which successfully extrapolated to larger clusters and discovered new stable silica structures for sizes n=28 to 51.
We present a new approach for the generation of stable structures of nanoclusters using deep learning methods. Our method consists in constructing an artificial potential energy surface, with local minima corresponding to the most stable structures and which is much smoother than "real" potential in the intermediate regions of the configuration space. To build the surface, graph convolutional networks are used. The method can extrapolates the potential surface to cases of structures with larger number of atoms than was used in training. Thus, having a sufficient number of low-energy structures in the training set, the method allows to generate new candidates for the ground-state structures, including ones with larger number of atoms. We applied the approach to silica clusters $(SiO_2)_n$ and for the first time found the stable structures with n=28...51. The method is universal and does not depend on the atomic composition and number of atoms.