Cargando…
Examining graph neural networks for crystal structures: Limitations and opportunities for capturing periodicity
Graph neural networks (GNNs) have recently been used to learn the representations of crystal structures through an end-to-end data-driven approach. However, a systematic top-down approach to evaluate and understand the limitations of GNNs in accurately capturing crystal structures has yet to be esta...
Autores principales: | , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
American Association for the Advancement of Science
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10637739/ https://www.ncbi.nlm.nih.gov/pubmed/37948518 http://dx.doi.org/10.1126/sciadv.adi3245 |
Sumario: | Graph neural networks (GNNs) have recently been used to learn the representations of crystal structures through an end-to-end data-driven approach. However, a systematic top-down approach to evaluate and understand the limitations of GNNs in accurately capturing crystal structures has yet to be established. In this study, we introduce an approach using human-designed descriptors as a compendium of human knowledge to investigate the extent to which GNNs can comprehend crystal structures. Our findings reveal that current state-of-the-art GNNs fall short in accurately capturing the periodicity of crystal structures. We analyze this failure by exploring three aspects: local expressive power, long-range information processing, and readout function. To address these identified limitations, we propose a straightforward and general solution: the hybridization of descriptors with GNNs, which directly supplements the missing information to GNNs. The hybridization enhances the predictive accuracy of GNNs for specific material properties, most notably phonon internal energy and heat capacity, which heavily rely on the periodicity of materials. |
---|