Xtal2DoS: Attention-based Crystal to Sequence Learning for Density of States Prediction
This addresses a challenging problem in materials science for researchers, focusing on spectral rather than scalar properties, with incremental improvements in speed and performance.
The paper tackled predicting spectral properties like density of states in materials science, proposing Xtal2DoS, an attention-based method that outperforms state-of-the-art models on four metrics for phonon and electronic DoS.
Modern machine learning techniques have been extensively applied to materials science, especially for property prediction tasks. A majority of these methods address scalar property predictions, while more challenging spectral properties remain less emphasized. We formulate a crystal-to-sequence learning task and propose a novel attention-based learning method, Xtal2DoS, which decodes the sequential representation of the material density of states (DoS) properties by incorporating the learned atomic embeddings through attention networks. Experiments show Xtal2DoS is faster than the existing models, and consistently outperforms other state-of-the-art methods on four metrics for two fundamental spectral properties, phonon and electronic DoS.