A Survey on Deep Semi-Supervised Learning

Deep semi-supervised learning is a fast-growing field with a range of practical applications. This paper provides a comprehensive survey on both fundamentals and recent advances in deep semi-supervised learning methods from perspectives of model design and unsupervised loss functions. We first prese...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on knowledge and data engineering Ročník 35; číslo 9; s. 8934 - 8954
Hlavní autoři: Yang, Xiangli, Song, Zixing, King, Irwin, Xu, Zenglin
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York IEEE 01.09.2023
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Data models
Deep learning
Deep semi-supervised learning
Regularization
Semi-supervised learning
Semisupervised learning
Supervised learning
Task analysis
Taxonomy
Training
ISSN:1041-4347, 1558-2191
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Deep semi-supervised learning is a fast-growing field with a range of practical applications. This paper provides a comprehensive survey on both fundamentals and recent advances in deep semi-supervised learning methods from perspectives of model design and unsupervised loss functions. We first present a taxonomy for deep semi-supervised learning that categorizes existing methods, including deep generative methods, consistency regularization methods, graph-based methods, pseudo-labeling methods, and hybrid methods. Then we provide a comprehensive review of 60 representative methods and offer a detailed comparison of these methods in terms of the type of losses, architecture differences, and test performance results. In addition to the progress in the past few years, we further discuss some shortcomings of existing methods and provide some tentative heuristic solutions for solving these open problems.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1041-4347
1558-2191
DOI:10.1109/TKDE.2022.3220219