Predicting Density of States via Multi-modal Transformer
This work addresses a domain-specific problem in materials science by providing a novel method for predicting DOS, which is incremental as it builds on transformer architectures for multi-modal data.
The paper tackles the problem of predicting the density of states (DOS) for materials by integrating crystal structure and energy information using a multi-modal transformer, achieving superior performance in experiments on Phonon and Electron DOS across real-world scenarios.
The density of states (DOS) is a spectral property of materials, which provides fundamental insights on various characteristics of materials. In this paper, we propose a model to predict the DOS by reflecting the nature of DOS: DOS determines the general distribution of states as a function of energy. Specifically, we integrate the heterogeneous information obtained from the crystal structure and the energies via multi-modal transformer, thereby modeling the complex relationships between the atoms in the crystal structure, and various energy levels. Extensive experiments on two types of DOS, i.e., Phonon DOS and Electron DOS, with various real-world scenarios demonstrate the superiority of DOSTransformer. The source code for DOSTransformer is available at https://github.com/HeewoongNoh/DOSTransformer.