Domain adaptive classification based on gradient weight pursuit

doi:10.13878/j.cnki.jnuist.20230927001

2025-6-21- 14

Home > Archive>Volume 17, Issue 2, 2025 >203-214. DOI:10.13878/j.cnki.jnuist.20230927001

Domain adaptive classification based on gradient weight pursuit
DOI:
                        10.13878/j.cnki.jnuist.20230927001
                    
CSTR:
                        
                    
Author:
                        CUI ShaojunCUI Shaojun
School of Automation, Nanjing University of Information Science & Technology, Nanjing 210044, China;Engineering Research Center of Digital Forensics Ministry of Education, Nanjing University of Information Science & Technology, Nanjing 210044, China;Jiangsu Key Laboratory of Big Data Analysis Technology, Nanjing University of Information Science & Technology, Nanjing 210044, China
Find this author on All Journals
Find this author on BaiDu
Search for this author on this site
JI FanfanJI Fanfan
Engineering Research Center of Digital Forensics Ministry of Education, Nanjing University of Information Science & Technology, Nanjing 210044, China;Jiangsu Key Laboratory of Big Data Analysis Technology, Nanjing University of Information Science & Technology, Nanjing 210044, China
Find this author on All Journals
Find this author on BaiDu
Search for this author on this site
WANG TingWANG Ting
Engineering Research Center of Digital Forensics Ministry of Education, Nanjing University of Information Science & Technology, Nanjing 210044, China;Jiangsu Key Laboratory of Big Data Analysis Technology, Nanjing University of Information Science & Technology, Nanjing 210044, China;School of Computer Science, Nanjing University of Information Science & Technology, Nanjing 210044, China
Find this author on All Journals
Find this author on BaiDu
Search for this author on this site
YUAN XiaotongYUAN Xiaotong
Engineering Research Center of Digital Forensics Ministry of Education, Nanjing University of Information Science & Technology, Nanjing 210044, China;Jiangsu Key Laboratory of Big Data Analysis Technology, Nanjing University of Information Science & Technology, Nanjing 210044, China;School of Computer Science, Nanjing University of Information Science & Technology, Nanjing 210044, China
Find this author on All Journals
Find this author on BaiDu
Search for this author on this site

                    
Affiliation:
Clc Number:TP391
Fund Project:

Article

Figures

Metrics

Reference

Cited by

Materials

Comments

Abstract:

Here,we propose a pruning and optimization approach based on Gradient Weight Pursuit (GWP) to address the overfitting in unsupervised domain,which manifests as significantly lower accuracy on downstream tasks compared to that on training sets.To tackle the overfitting challenge in unsupervised domain,we employ the dense-sparse-dense strategy,focusing on both difference-based and adversarial adaptive methods.First,the network is pretrained intensively to identify crucial connections.Second,during the pruning stage,the optimization algorithm in this paper distinguishes itself from original dense-sparse-dense strategy by jointly considering both weight and gradient information.Specifically,it leverages both weight (i.e.zero-order information) and gradient (i.e.first-order information) to influence pruning process.In the final dense phase,the pruned connections are restored and the dense network is retrained with a reduced learning rate.Finally,the obtained network achieves desirable outcomes in downstream tasks.The experimental results show that the proposed GWP approach can effectively improve the accuracy of downstream tasks,offering a plug-and-play capability compared with original difference-based and adversarial domain adaptation methods.

Key words:gradient weight pursuit (GWP);unsupervised domain adaptation(UDA);dense-sparse-dense (DSD);overfitting;zero-order information;first-order information

Get Citation

CUI Shaojun, JI Fanfan, WANG Ting, YUAN Xiaotong. Domain adaptive classification based on gradient weight pursuit[J]. Journal of Nanjing University of Information Science & Technology,2025,17(2):203-214

Copy

Article Metrics

Abstract:
PDF:
HTML:
Cited by:

History

Received:September 27,2023
Revised:
Adopted:
Online: April 16,2025
Published: March 28,2025

Article QR Code

Address：No. 219, Ningliu Road, Nanjing, Jiangsu Province

Postcode：210044

Phone：025-58731025

Get Citation

Related Videos

Share

Article Metrics

History

Article QR Code