Automated stopping criterion for spectral measurements with active learning

Abstract The automated stopping of a spectral measurement with active learning is proposed. The optimal stopping of the measurement is realised with a stopping criterion based on the upper bound of the posterior average of the generalisation error of the Gaussian process regression. It is revealed t...

Full description

Bibliographic Details
Main Authors:	Tetsuro Ueno, Hideaki Ishibashi, Hideitsu Hino, Kanta Ono
Format:	Article
Language:	English
Published:	Nature Publishing Group 2021-08-01
Series:	npj Computational Materials
Online Access:	https://doi.org/10.1038/s41524-021-00606-5

id	doaj-7d0e960186a94c6d849eea07d0d17fa7
record_format	Article
spelling	doaj-7d0e960186a94c6d849eea07d0d17fa72021-08-29T11:18:09ZengNature Publishing Groupnpj Computational Materials2057-39602021-08-01711910.1038/s41524-021-00606-5Automated stopping criterion for spectral measurements with active learningTetsuro Ueno0Hideaki Ishibashi1Hideitsu Hino2Kanta Ono3Synchrotron Radiation Research Center, Kansai Photon Science Institute, Quantum Beam Science Research Directorate, National Institutes for Quantum and Radiological Science and TechnologyDepartment of Human Intelligence Systems, Graduate School of Life Science and Systems Engineering, Kyushu Institute of TechnologyThe Institute of Statistical Mathematics, Research Organization of Information and SystemsDepartment of Applied Physics, Graduate School of Engineering, Osaka UniversityAbstract The automated stopping of a spectral measurement with active learning is proposed. The optimal stopping of the measurement is realised with a stopping criterion based on the upper bound of the posterior average of the generalisation error of the Gaussian process regression. It is revealed that the automated stopping criterion of the spectral measurement gives an approximated X-ray absorption spectrum with sufficient accuracy and reduced data size. The proposed method is not only a proof-of-concept of the optimal stopping problem in active learning but also the key to enhancing the efficiency of spectral measurements for high-throughput experiments in the era of materials informatics.https://doi.org/10.1038/s41524-021-00606-5
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Tetsuro Ueno Hideaki Ishibashi Hideitsu Hino Kanta Ono
spellingShingle	Tetsuro Ueno Hideaki Ishibashi Hideitsu Hino Kanta Ono Automated stopping criterion for spectral measurements with active learning npj Computational Materials
author_facet	Tetsuro Ueno Hideaki Ishibashi Hideitsu Hino Kanta Ono
author_sort	Tetsuro Ueno
title	Automated stopping criterion for spectral measurements with active learning
title_short	Automated stopping criterion for spectral measurements with active learning
title_full	Automated stopping criterion for spectral measurements with active learning
title_fullStr	Automated stopping criterion for spectral measurements with active learning
title_full_unstemmed	Automated stopping criterion for spectral measurements with active learning
title_sort	automated stopping criterion for spectral measurements with active learning
publisher	Nature Publishing Group
series	npj Computational Materials
issn	2057-3960
publishDate	2021-08-01
description	Abstract The automated stopping of a spectral measurement with active learning is proposed. The optimal stopping of the measurement is realised with a stopping criterion based on the upper bound of the posterior average of the generalisation error of the Gaussian process regression. It is revealed that the automated stopping criterion of the spectral measurement gives an approximated X-ray absorption spectrum with sufficient accuracy and reduced data size. The proposed method is not only a proof-of-concept of the optimal stopping problem in active learning but also the key to enhancing the efficiency of spectral measurements for high-throughput experiments in the era of materials informatics.
url	https://doi.org/10.1038/s41524-021-00606-5
work_keys_str_mv	AT tetsuroueno automatedstoppingcriterionforspectralmeasurementswithactivelearning AT hideakiishibashi automatedstoppingcriterionforspectralmeasurementswithactivelearning AT hideitsuhino automatedstoppingcriterionforspectralmeasurementswithactivelearning AT kantaono automatedstoppingcriterionforspectralmeasurementswithactivelearning
_version_	1721186939956101120

Automated stopping criterion for spectral measurements with active learning

Similar Items