Machine learning identifies scale-free properties in disordered materials
The performance of a trained neural network may be biased even by generic features of its architecture. Yu et al. ask for the disordered lattice of atoms producing a certain wave localization and the network prefers to answer with power-law distributed displacements.
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2020-09-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-020-18653-9 |
| _version_ | 1818683183207546880 |
|---|---|
| author | Sunkyu Yu Xianji Piao Namkyoo Park |
| author_facet | Sunkyu Yu Xianji Piao Namkyoo Park |
| author_sort | Sunkyu Yu |
| collection | DOAJ |
| description | The performance of a trained neural network may be biased even by generic features of its architecture. Yu et al. ask for the disordered lattice of atoms producing a certain wave localization and the network prefers to answer with power-law distributed displacements. |
| first_indexed | 2024-12-17T10:30:41Z |
| format | Article |
| id | doaj.art-e0f5eff1660b4e30a8c655466ff74960 |
| institution | Directory Open Access Journal |
| issn | 2041-1723 |
| language | English |
| last_indexed | 2024-12-17T10:30:41Z |
| publishDate | 2020-09-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj.art-e0f5eff1660b4e30a8c655466ff749602022-12-21T21:52:33ZengNature PortfolioNature Communications2041-17232020-09-0111111110.1038/s41467-020-18653-9Machine learning identifies scale-free properties in disordered materialsSunkyu Yu0Xianji Piao1Namkyoo Park2Photonic Systems Laboratory, Department of Electrical and Computer Engineering, Seoul National UniversityPhotonic Systems Laboratory, Department of Electrical and Computer Engineering, Seoul National UniversityPhotonic Systems Laboratory, Department of Electrical and Computer Engineering, Seoul National UniversityThe performance of a trained neural network may be biased even by generic features of its architecture. Yu et al. ask for the disordered lattice of atoms producing a certain wave localization and the network prefers to answer with power-law distributed displacements.https://doi.org/10.1038/s41467-020-18653-9 |
| spellingShingle | Sunkyu Yu Xianji Piao Namkyoo Park Machine learning identifies scale-free properties in disordered materials Nature Communications |
| title | Machine learning identifies scale-free properties in disordered materials |
| title_full | Machine learning identifies scale-free properties in disordered materials |
| title_fullStr | Machine learning identifies scale-free properties in disordered materials |
| title_full_unstemmed | Machine learning identifies scale-free properties in disordered materials |
| title_short | Machine learning identifies scale-free properties in disordered materials |
| title_sort | machine learning identifies scale free properties in disordered materials |
| url | https://doi.org/10.1038/s41467-020-18653-9 |
| work_keys_str_mv | AT sunkyuyu machinelearningidentifiesscalefreepropertiesindisorderedmaterials AT xianjipiao machinelearningidentifiesscalefreepropertiesindisorderedmaterials AT namkyoopark machinelearningidentifiesscalefreepropertiesindisorderedmaterials |