A multi-task learning model for global soil moisture prediction based on adaptive weight allocation

Abstract Global soil moisture is crucial for multiple disciplines such as earth science and agricultural science. Therefore, there are many studies on how to improve the prediction accuracy of soil moisture, especially the rapid development of deep learning in recent years, which has greatly improve...

Full description

Saved in:
Bibliographic Details
Main Authors: Yeguang li, Haiying Liu, Tianshan Lv
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-01894-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850243774916067328
author Yeguang li
Haiying Liu
Tianshan Lv
author_facet Yeguang li
Haiying Liu
Tianshan Lv
author_sort Yeguang li
collection DOAJ
description Abstract Global soil moisture is crucial for multiple disciplines such as earth science and agricultural science. Therefore, there are many studies on how to improve the prediction accuracy of soil moisture, especially the rapid development of deep learning in recent years, which has greatly improved data-driven models. This paper proposes an adaptive weight long short-term memory (AW-LSTM) model based on dynamic weight allocation, which dynamically optimizes the model by calculating the correlation coefficient $$p$$ between tasks. The experimental verification demonstrated that the AW-LSTM model exhibited the most accurate soil moisture prediction, with $$R$$ and $$R^2$$ values of 0.9456 and 0.8489, respectively. These values were 0.019 and 0.072 higher than the single-task prediction values observed in the benchmark dataset.
format Article
id doaj-art-190d88fcb2b546e681dce45874790449
institution OA Journals
issn 2045-2322
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-190d88fcb2b546e681dce458747904492025-08-20T01:59:56ZengNature PortfolioScientific Reports2045-23222025-05-0115111310.1038/s41598-025-01894-3A multi-task learning model for global soil moisture prediction based on adaptive weight allocationYeguang li0Haiying Liu1Tianshan Lv2School of Economics and Management, Changchun University of TechnologySchool of Economics and Management, Changchun University of TechnologySchool of Economics and Management, Changchun University of TechnologyAbstract Global soil moisture is crucial for multiple disciplines such as earth science and agricultural science. Therefore, there are many studies on how to improve the prediction accuracy of soil moisture, especially the rapid development of deep learning in recent years, which has greatly improved data-driven models. This paper proposes an adaptive weight long short-term memory (AW-LSTM) model based on dynamic weight allocation, which dynamically optimizes the model by calculating the correlation coefficient $$p$$ between tasks. The experimental verification demonstrated that the AW-LSTM model exhibited the most accurate soil moisture prediction, with $$R$$ and $$R^2$$ values of 0.9456 and 0.8489, respectively. These values were 0.019 and 0.072 higher than the single-task prediction values observed in the benchmark dataset.https://doi.org/10.1038/s41598-025-01894-3Multi-task learningSoil moistureDeep learning
spellingShingle Yeguang li
Haiying Liu
Tianshan Lv
A multi-task learning model for global soil moisture prediction based on adaptive weight allocation
Scientific Reports
Multi-task learning
Soil moisture
Deep learning
title A multi-task learning model for global soil moisture prediction based on adaptive weight allocation
title_full A multi-task learning model for global soil moisture prediction based on adaptive weight allocation
title_fullStr A multi-task learning model for global soil moisture prediction based on adaptive weight allocation
title_full_unstemmed A multi-task learning model for global soil moisture prediction based on adaptive weight allocation
title_short A multi-task learning model for global soil moisture prediction based on adaptive weight allocation
title_sort multi task learning model for global soil moisture prediction based on adaptive weight allocation
topic Multi-task learning
Soil moisture
Deep learning
url https://doi.org/10.1038/s41598-025-01894-3
work_keys_str_mv AT yeguangli amultitasklearningmodelforglobalsoilmoisturepredictionbasedonadaptiveweightallocation
AT haiyingliu amultitasklearningmodelforglobalsoilmoisturepredictionbasedonadaptiveweightallocation
AT tianshanlv amultitasklearningmodelforglobalsoilmoisturepredictionbasedonadaptiveweightallocation
AT yeguangli multitasklearningmodelforglobalsoilmoisturepredictionbasedonadaptiveweightallocation
AT haiyingliu multitasklearningmodelforglobalsoilmoisturepredictionbasedonadaptiveweightallocation
AT tianshanlv multitasklearningmodelforglobalsoilmoisturepredictionbasedonadaptiveweightallocation