Xtal2DoS: Attention-based Crystal to Sequence Learning for Density of States PredictionDownload PDF

Published: 21 Oct 2022, Last Modified: 03 Jul 2024AI4Science PosterReaders: Everyone
Keywords: scientific discovery, sequence learning, GNN, attention, transformer, machine learning
Abstract: 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.
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](https://www.catalyzex.com/paper/xtal2dos-attention-based-crystal-to-sequence/code)
0 Replies

Loading