Next Article in Journal
YOLO-MFD: Object Detection for Multi-Scenario Fires
Previous Article in Journal
YOLO-SSFA: A Lightweight Real-Time Infrared Detection Method for Small Targets
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Information
Volume 16
Issue 7
10.3390/info16070619
information-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Adaptive Multi-Gradient Guidance with Conflict Resolution for Limited-Sample Regression

by
Yu Lin
1,
Jiaxiang Lin
1,*,
Keju Zhang
1,
Qin Zheng
1,
Liqiang Lin
2 and
Qianqian Chen
1
1
Key Laboratory of Smart Agriculture and Forestry, Fujian Agriculture and Forestry University, Fuzhou 350002, China
2
Center for Agroforestry Mega Data Science, Fujian Agriculture and Forestry University, Fuzhou 350002, China
*
Author to whom correspondence should be addressed.
Information 2025, 16(7), 619; https://doi.org/10.3390/info16070619
Submission received: 16 June 2025 / Revised: 14 July 2025 / Accepted: 17 July 2025 / Published: 21 July 2025
(This article belongs to the Section Artificial Intelligence)
Browse Figures
Versions Notes

Abstract

Recent studies report that gradient guidance extracted from a single-reference model can improve Limited-Sample regression. However, one reference model may not capture all relevant characteristics of the target function, which can restrict the capacity of the learner. To address this issue, we introduce the Multi-Gradient Guided Network (MGGN), an extension of single-gradient guidance that combines gradients from several reference models. The gradients are merged through an adaptive weighting scheme, and an orthogonal-projection step is applied to reduce potential conflicts between them. Experiments on sine regression are used to evaluate the method. The results indicate that MGGN achieves higher predictive accuracy and improved stability than existing single-gradient guidance and meta-learning baselines, benefiting from the complementary information provided by multiple reference models.
Keywords: multi-gradient fusion; Limited-Sample regression; PCGrad; neural networks multi-gradient fusion; Limited-Sample regression; PCGrad; neural networks

Share and Cite

MDPI and ACS Style

Lin, Y.; Lin, J.; Zhang, K.; Zheng, Q.; Lin, L.; Chen, Q. Adaptive Multi-Gradient Guidance with Conflict Resolution for Limited-Sample Regression. Information 2025, 16, 619. https://doi.org/10.3390/info16070619

AMA Style

Lin Y, Lin J, Zhang K, Zheng Q, Lin L, Chen Q. Adaptive Multi-Gradient Guidance with Conflict Resolution for Limited-Sample Regression. Information. 2025; 16(7):619. https://doi.org/10.3390/info16070619

Chicago/Turabian Style

Lin, Yu, Jiaxiang Lin, Keju Zhang, Qin Zheng, Liqiang Lin, and Qianqian Chen. 2025. "Adaptive Multi-Gradient Guidance with Conflict Resolution for Limited-Sample Regression" Information 16, no. 7: 619. https://doi.org/10.3390/info16070619

APA Style

Lin, Y., Lin, J., Zhang, K., Zheng, Q., Lin, L., & Chen, Q. (2025). Adaptive Multi-Gradient Guidance with Conflict Resolution for Limited-Sample Regression. Information, 16(7), 619. https://doi.org/10.3390/info16070619

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop