| Autor: |
Speckhard D; Physics Department and CSMB, Humboldt-Universität zu Berlin, Zum Großen Windkanal 2, 12489 Berlin, Germany. claudia.draxl@physik.hu-berlin.de.; Max Planck Institute for Solid State Research, Heisenbergstraaae 1, 70569 Stuttgart, Germany., Bechtel T; Physics Department and CSMB, Humboldt-Universität zu Berlin, Zum Großen Windkanal 2, 12489 Berlin, Germany. claudia.draxl@physik.hu-berlin.de.; Max Planck Institute for Solid State Research, Heisenbergstraaae 1, 70569 Stuttgart, Germany., Ghiringhelli LM; Department of Materials Science and Engineering, Friedrich-Alexander Universität Erlangen-Nürnberg, Dr.-Mack-Str. 77, 90762 Fürth, Germany., Kuban M; Physics Department and CSMB, Humboldt-Universität zu Berlin, Zum Großen Windkanal 2, 12489 Berlin, Germany. claudia.draxl@physik.hu-berlin.de., Rigamonti S; Physics Department and CSMB, Humboldt-Universität zu Berlin, Zum Großen Windkanal 2, 12489 Berlin, Germany. claudia.draxl@physik.hu-berlin.de., Draxl C; Physics Department and CSMB, Humboldt-Universität zu Berlin, Zum Großen Windkanal 2, 12489 Berlin, Germany. claudia.draxl@physik.hu-berlin.de.; Max Planck Institute for Solid State Research, Heisenbergstraaae 1, 70569 Stuttgart, Germany. |
| Abstrakt: |
Big data has ushered in a new wave of predictive power using machine-learning models. In this work, we assess what big means in the context of typical materials-science machine-learning problems. This concerns not only data volume, but also data quality and veracity as much as infrastructure issues. With selected examples, we ask (i) how models generalize to similar datasets, (ii) how high-quality datasets can be gathered from heterogenous sources, (iii) how the feature set and complexity of a model can affect expressivity, and (iv) what infrastructure requirements are needed to create larger datasets and train models on them. In sum, we find that big data present unique challenges along very different aspects that should serve to motivate further work. |
