Method for Knowledge Transfer via Multi-Task Semi-Supervised Self-Paced

Adequate labeled data is essential for learning a reliable and generalizable model in many machine learning tasks. However, labeled data is becoming scarce and costly to obtain, which has spurred consistent interest in knowledge transfer techniques. Therefore, semi-supervised and multi-task learning...

Full description

Saved in:
Bibliographic Details
Main Authors: Yao Zhao, Hongying Liu, Huaxian Pan, Zhen Song, Chunting Liu, Anni Wei, Baoshuang Zhang, Wei Lu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11017642/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850219269394006016
author Yao Zhao
Hongying Liu
Huaxian Pan
Zhen Song
Chunting Liu
Anni Wei
Baoshuang Zhang
Wei Lu
author_facet Yao Zhao
Hongying Liu
Huaxian Pan
Zhen Song
Chunting Liu
Anni Wei
Baoshuang Zhang
Wei Lu
author_sort Yao Zhao
collection DOAJ
description Adequate labeled data is essential for learning a reliable and generalizable model in many machine learning tasks. However, labeled data is becoming scarce and costly to obtain, which has spurred consistent interest in knowledge transfer techniques. Therefore, semi-supervised and multi-task learning is combined to alleviate the challenge, but the complexity of the task should be considered. To achieve more effective knowledge transfer with limited labeled data, we propose a unified multi-task semi-supervised self-paced learning (MSSP) scheme in this paper. MSSP naturally integrates the common structures shared by multiple related tasks and the manifold structure regularized by unlabeled data, enabling the respective knowledge transferred from the feature space and the instance space to complement and constrain each other. This leads to faster and more accurate searches in the underlying hypothesis space. We adopt Alternating convex search (ACS) method to solve MSSP, that is, each iteration sequentially trains the prediction model with a fixed set of labeled instances and then updates the labeled training set by adding more complex instances. With the aid of a self-controlled learning pace, a more robust and globally optimal model can be gradually constructed. Experimental results on several benchmark datasets show that our method achieves a performance gain of 3%-15% in classification accuracy compared to baseline algorithms, along with significant advantages in convergence speed.
format Article
id doaj-art-99471709e7964bda9ec196cf8a9ef361
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-99471709e7964bda9ec196cf8a9ef3612025-08-20T02:07:26ZengIEEEIEEE Access2169-35362025-01-011310140510141410.1109/ACCESS.2025.357498211017642Method for Knowledge Transfer via Multi-Task Semi-Supervised Self-PacedYao Zhao0https://orcid.org/0009-0008-4302-3663Hongying Liu1Huaxian Pan2https://orcid.org/0009-0006-9505-8097Zhen Song3Chunting Liu4Anni Wei5https://orcid.org/0009-0008-9449-8722Baoshuang Zhang6https://orcid.org/0000-0002-4166-6568Wei Lu7https://orcid.org/0000-0002-0098-7584School of Economics and Statistics, Xingzhi College, Xi’an University of Finance and Economics, Xi’an, Shaanxi, ChinaSchool of Economics and Statistics, Xingzhi College, Xi’an University of Finance and Economics, Xi’an, Shaanxi, ChinaSchool of Economics and Statistics, Xingzhi College, Xi’an University of Finance and Economics, Xi’an, Shaanxi, ChinaSchool of Economics and Statistics, Xingzhi College, Xi’an University of Finance and Economics, Xi’an, Shaanxi, ChinaSchool of Economics and Statistics, Xingzhi College, Xi’an University of Finance and Economics, Xi’an, Shaanxi, ChinaSchool of Economics and Statistics, Xingzhi College, Xi’an University of Finance and Economics, Xi’an, Shaanxi, ChinaSchool of Economics and Statistics, Xingzhi College, Xi’an University of Finance and Economics, Xi’an, Shaanxi, ChinaSchool of Information, Xi’an University of Finance and Economics, Xi’an, Shaanxi, ChinaAdequate labeled data is essential for learning a reliable and generalizable model in many machine learning tasks. However, labeled data is becoming scarce and costly to obtain, which has spurred consistent interest in knowledge transfer techniques. Therefore, semi-supervised and multi-task learning is combined to alleviate the challenge, but the complexity of the task should be considered. To achieve more effective knowledge transfer with limited labeled data, we propose a unified multi-task semi-supervised self-paced learning (MSSP) scheme in this paper. MSSP naturally integrates the common structures shared by multiple related tasks and the manifold structure regularized by unlabeled data, enabling the respective knowledge transferred from the feature space and the instance space to complement and constrain each other. This leads to faster and more accurate searches in the underlying hypothesis space. We adopt Alternating convex search (ACS) method to solve MSSP, that is, each iteration sequentially trains the prediction model with a fixed set of labeled instances and then updates the labeled training set by adding more complex instances. With the aid of a self-controlled learning pace, a more robust and globally optimal model can be gradually constructed. Experimental results on several benchmark datasets show that our method achieves a performance gain of 3%-15% in classification accuracy compared to baseline algorithms, along with significant advantages in convergence speed.https://ieeexplore.ieee.org/document/11017642/Multi-task learningself-paced learningsemi-supervised learningalternating convex search
spellingShingle Yao Zhao
Hongying Liu
Huaxian Pan
Zhen Song
Chunting Liu
Anni Wei
Baoshuang Zhang
Wei Lu
Method for Knowledge Transfer via Multi-Task Semi-Supervised Self-Paced
IEEE Access
Multi-task learning
self-paced learning
semi-supervised learning
alternating convex search
title Method for Knowledge Transfer via Multi-Task Semi-Supervised Self-Paced
title_full Method for Knowledge Transfer via Multi-Task Semi-Supervised Self-Paced
title_fullStr Method for Knowledge Transfer via Multi-Task Semi-Supervised Self-Paced
title_full_unstemmed Method for Knowledge Transfer via Multi-Task Semi-Supervised Self-Paced
title_short Method for Knowledge Transfer via Multi-Task Semi-Supervised Self-Paced
title_sort method for knowledge transfer via multi task semi supervised self paced
topic Multi-task learning
self-paced learning
semi-supervised learning
alternating convex search
url https://ieeexplore.ieee.org/document/11017642/
work_keys_str_mv AT yaozhao methodforknowledgetransferviamultitasksemisupervisedselfpaced
AT hongyingliu methodforknowledgetransferviamultitasksemisupervisedselfpaced
AT huaxianpan methodforknowledgetransferviamultitasksemisupervisedselfpaced
AT zhensong methodforknowledgetransferviamultitasksemisupervisedselfpaced
AT chuntingliu methodforknowledgetransferviamultitasksemisupervisedselfpaced
AT anniwei methodforknowledgetransferviamultitasksemisupervisedselfpaced
AT baoshuangzhang methodforknowledgetransferviamultitasksemisupervisedselfpaced
AT weilu methodforknowledgetransferviamultitasksemisupervisedselfpaced