Source-free unsupervised domain adaptation: A survey

Unsupervised domain adaptation (UDA) via deep learning has attracted appealing attention for tackling domain-shift problems caused by distribution discrepancy across different domains. Existing UDA approaches highly depend on the accessibility of source domain data, which is usually limited in pract...

Full description

Saved in:
Bibliographic Details
Published in:Neural networks Vol. 174; p. 106230
Main Authors: Fang, Yuqi, Yap, Pew-Thian, Lin, Weili, Zhu, Hongtu, Liu, Mingxia
Format: Journal Article
Language:English
Published: United States Elsevier Ltd 01.06.2024
Subjects:
Deep Learning
Domain adaptation
Humans
Neural Networks, Computer
Source-free
Survey
Unsupervised learning
Unsupervised Machine Learning
Domain adaptation
Source-free
Survey
Unsupervised learning
ISSN:0893-6080, 1879-2782, 1879-2782
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Unsupervised domain adaptation (UDA) via deep learning has attracted appealing attention for tackling domain-shift problems caused by distribution discrepancy across different domains. Existing UDA approaches highly depend on the accessibility of source domain data, which is usually limited in practical scenarios due to privacy protection, data storage and transmission cost, and computation burden. To tackle this issue, many source-free unsupervised domain adaptation (SFUDA) methods have been proposed recently, which perform knowledge transfer from a pre-trained source model to the unlabeled target domain with source data inaccessible. A comprehensive review of these works on SFUDA is of great significance. In this paper, we provide a timely and systematic literature review of existing SFUDA approaches from a technical perspective. Specifically, we categorize current SFUDA studies into two groups, i.e., white-box SFUDA and black-box SFUDA, and further divide them into finer subcategories based on different learning strategies they use. We also investigate the challenges of methods in each subcategory, discuss the advantages/disadvantages of white-box and black-box SFUDA methods, conclude the commonly used benchmark datasets, and summarize the popular techniques for improved generalizability of models learned without using source data. We finally discuss several promising future directions in this field. •Providing a review of existing source-free unsupervised domain adaptation approaches.•Dividing current methods into finer categories and discussing their (dis)advantages.•Summarizing benchmark datasets and techniques that improve model generalizability.•Discussing promising research directions to guide trajectory of future investigation.•Offering extensive references, serving as a valuable resource in this research field.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Undefined-1
ObjectType-Feature-3
ObjectType-Review-4
content type line 23
ISSN:0893-6080
1879-2782
1879-2782
DOI:10.1016/j.neunet.2024.106230