Auto-MatRegressor: liberating machine learning alchemists

Simple item page
dc.contributor.authorLiu, Yen_AU
dc.contributor.authorWang, Sen_AU
dc.contributor.authorYang, Zen_AU
dc.contributor.authorAvdeev, Men_AU
dc.contributor.authorShi, Sen_AU
dc.date.accessioned2024-12-05T03:08:38Zen_AU
dc.date.available2024-12-05T03:08:38Zen_AU
dc.date.issued2023-06-30en_AU
dc.date.statistics2024-06-13en_AU
dc.description.abstractMachine learning (ML) is widely used to uncover structure–property relationships of materials due to its ability to quickly find potential data patterns and make accurate predictions. However, like alchemists, materials scientists are plagued by time-consuming and labor-intensive experiments to build high-accuracy ML models. Here, we propose an automatic modeling method based on meta-learning for materials property prediction named Auto-MatRegressor, which automates algorithm selection and hyperparameter optimization by learning from previous modeling experience, i.e., meta-data on historical datasets. The meta-data used in this work consists of 27 meta-features that characterize the datasets and the prediction performances of 18 algorithms commonly used in materials science. To recommend optimal algorithms, a collaborative meta-learning method embedded with domain knowledge quantified by a materials categories tree is designed. Experiments on 60 datasets show that compared with the traditional modeling method from scratch, Auto-MatRegressor automatically selects appropriate algorithms at lower computational cost, which accelerates constructing ML models with good prediction accuracy. Auto-MatRegressor supports dynamic expansion of meta-data with the increase of the number of materials datasets and other required algorithms and can be applied to any ML materials discovery and design task. © 2023 Science China Press. Published by Elsevier B.V. and Science China Press. All rights reserved.en_AU
dc.format.mediumPrint-Electronicen_AU
dc.identifier.citationLiu, Y., Wang, S., Yang, Z., Avdeev, M., & Shi, S. (2023). Auto-MatRegressor: liberating machine learning alchemists. Science Bulletin, 68(12), 1259-1270. doi.:10.1016/j.scib.2023.05.017en_AU
dc.identifier.issn2095-9273en_AU
dc.identifier.issn2095-9281en_AU
dc.identifier.issue12en_AU
dc.identifier.journaltitleScience Bulletinen_AU
dc.identifier.pagination1259-1270en_AU
dc.identifier.urihttps://doi.org/10.1016/j.scib.2023.05.017en_AU
dc.identifier.urihttps://apo.ansto.gov.au/handle/10238/15772en_AU
dc.identifier.volume68en_AU
dc.languageEnglishen_AU
dc.language.isoenen_AU
dc.publisherElsevieren_AU
dc.subjectMachine Learningen_AU
dc.subjectMaterials testingen_AU
dc.subjectDatasetsen_AU
dc.subjectAlgorithmsen_AU
dc.subjectAutomationen_AU
dc.subjectPrediction equationsen_AU
dc.subjectMaterialsen_AU
dc.titleAuto-MatRegressor: liberating machine learning alchemistsen_AU
dc.typeJournal Articleen_AU
dcterms.dateAccepted2023-05-08en_AU
Files
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.66 KB
Format:
Plain Text
Description:
Download
Collections
Journal Articles