dc.identifier.uri |
http://dx.doi.org/10.15488/13094 |
|
dc.identifier.uri |
https://www.repo.uni-hannover.de/handle/123456789/13199 |
|
dc.contributor.author |
Das, Kishalay
|
|
dc.contributor.author |
Samanta, Bidisha
|
|
dc.contributor.author |
Goyal, Pawan
|
|
dc.contributor.author |
Lee, Seung-Cheol
|
|
dc.contributor.author |
Bhattacharjee, Satadeep
|
|
dc.contributor.author |
Ganguly, Niloy
|
|
dc.date.accessioned |
2022-12-06T12:06:45Z |
|
dc.date.available |
2022-12-06T12:06:45Z |
|
dc.date.issued |
2022 |
|
dc.identifier.citation |
Das, K.; Samanta, B.; Goyal, P.; Lee, S.-C.; Bhattacharjee, S. et al.: CrysXPP: An explainable property predictor for crystalline materials. In: npj Computational Materials 8 (2022), 43. DOI: https://doi.org/10.1038/s41524-022-00716-8 |
|
dc.description.abstract |
We present a deep-learning framework, CrysXPP, to allow rapid and accurate prediction of electronic, magnetic, and elastic properties of a wide range of materials. CrysXPP lowers the need for large property tagged datasets by intelligently designing an autoencoder, CrysAE. The important structural and chemical properties captured by CrysAE from a large amount of available crystal graphs data helped in achieving low prediction errors. Moreover, we design a feature selector that helps to interpret the model’s prediction. Most notably, when given a small amount of experimental data, CrysXPP is consistently able to outperform conventional DFT. A detailed ablation study establishes the importance of different design steps. We release the large pre-trained model CrysAE. We believe by fine-tuning the model with a small amount of property-tagged data, researchers can achieve superior performance on various applications with a restricted data source. |
eng |
dc.language.iso |
eng |
|
dc.publisher |
London : Nature Publ. Group |
|
dc.relation.ispartofseries |
npj Computational Materials 8 (2022) |
|
dc.rights |
CC BY 4.0 Unported |
|
dc.rights.uri |
https://creativecommons.org/licenses/by/4.0/ |
|
dc.subject |
Chromium compounds |
eng |
dc.subject |
Design for testability |
eng |
dc.subject |
Forecasting |
eng |
dc.subject |
Large dataset |
eng |
dc.subject |
Accurate prediction |
eng |
dc.subject.ddc |
004 | Informatik
|
ger |
dc.title |
CrysXPP: An explainable property predictor for crystalline materials |
eng |
dc.type |
Article |
|
dc.type |
Text |
|
dc.relation.essn |
2057-3960 |
|
dc.relation.doi |
https://doi.org/10.1038/s41524-022-00716-8 |
|
dc.bibliographicCitation.volume |
8 |
|
dc.bibliographicCitation.firstPage |
43 |
|
dc.description.version |
publishedVersion |
|
tib.accessRights |
frei zug�nglich |
|