Home > Research > Publications & Outputs > Trusted Semi-Supervised Multi-View Classificati...

Links

Text available via DOI:

View graph of relations

Trusted Semi-Supervised Multi-View Classification With Contrastive Learning

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Trusted Semi-Supervised Multi-View Classification With Contrastive Learning. / Wang, Xiaoli; Wang, Yongli; Wang, Yupeng et al.
In: IEEE Transactions on Multimedia, Vol. 26, 31.12.2024, p. 8268-8278.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Wang, X, Wang, Y, Wang, Y, Huang, A & Liu, J 2024, 'Trusted Semi-Supervised Multi-View Classification With Contrastive Learning', IEEE Transactions on Multimedia, vol. 26, pp. 8268-8278. https://doi.org/10.1109/TMM.2024.3379079

APA

Wang, X., Wang, Y., Wang, Y., Huang, A., & Liu, J. (2024). Trusted Semi-Supervised Multi-View Classification With Contrastive Learning. IEEE Transactions on Multimedia, 26, 8268-8278. https://doi.org/10.1109/TMM.2024.3379079

Vancouver

Wang X, Wang Y, Wang Y, Huang A, Liu J. Trusted Semi-Supervised Multi-View Classification With Contrastive Learning. IEEE Transactions on Multimedia. 2024 Dec 31;26:8268-8278. Epub 2024 Mar 19. doi: 10.1109/TMM.2024.3379079

Author

Wang, Xiaoli ; Wang, Yongli ; Wang, Yupeng et al. / Trusted Semi-Supervised Multi-View Classification With Contrastive Learning. In: IEEE Transactions on Multimedia. 2024 ; Vol. 26. pp. 8268-8278.

Bibtex

@article{9e8b6b501d5f400b927225fa44dd2aa4,
title = "Trusted Semi-Supervised Multi-View Classification With Contrastive Learning",
abstract = "Semi-supervised multi-view learning is a remarkable but challenging task. Existing semi-supervised multi-view classification (SMVC) approaches mainly focus on performance improvement while ignoring decision reliability, which limits their deployment in safety-critical applications. Although several trusted multi-view classification methods are proposed recently, they rely on manual annotations. Therefore, this work emphasizes trusted multi-view classification learning under semi-supervised conditions. Different from existing SMVC methods, this work jointly models class probabilities and uncertainties based on evidential deep learning to formulate view-specific opinions. Moreover, unlike previous works that explore cross-view consistency in a single schema, this work proposes a multi-level consistency constraint. Specifically, we explore instance-level consistency on the view-specific representation space and category-level consistency on opinions from multiple views. Our proposed trusted graph-based contrastive loss nicely establishes the relationship between joint opinions and view-specific representations, which enables view-specific representations to enjoy a good manifold to improve classification performance. Overall, the proposed approach provides reliable and superior semi-supervised multi-view classification decisions. Extensive experiments demonstrate the effectiveness, reliability and robustness of the proposed model.",
author = "Xiaoli Wang and Yongli Wang and Yupeng Wang and Anqi Huang and Jun Liu",
year = "2024",
month = dec,
day = "31",
doi = "10.1109/TMM.2024.3379079",
language = "English",
volume = "26",
pages = "8268--8278",
journal = "IEEE Transactions on Multimedia",
issn = "1520-9210",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

RIS

TY - JOUR

T1 - Trusted Semi-Supervised Multi-View Classification With Contrastive Learning

AU - Wang, Xiaoli

AU - Wang, Yongli

AU - Wang, Yupeng

AU - Huang, Anqi

AU - Liu, Jun

PY - 2024/12/31

Y1 - 2024/12/31

N2 - Semi-supervised multi-view learning is a remarkable but challenging task. Existing semi-supervised multi-view classification (SMVC) approaches mainly focus on performance improvement while ignoring decision reliability, which limits their deployment in safety-critical applications. Although several trusted multi-view classification methods are proposed recently, they rely on manual annotations. Therefore, this work emphasizes trusted multi-view classification learning under semi-supervised conditions. Different from existing SMVC methods, this work jointly models class probabilities and uncertainties based on evidential deep learning to formulate view-specific opinions. Moreover, unlike previous works that explore cross-view consistency in a single schema, this work proposes a multi-level consistency constraint. Specifically, we explore instance-level consistency on the view-specific representation space and category-level consistency on opinions from multiple views. Our proposed trusted graph-based contrastive loss nicely establishes the relationship between joint opinions and view-specific representations, which enables view-specific representations to enjoy a good manifold to improve classification performance. Overall, the proposed approach provides reliable and superior semi-supervised multi-view classification decisions. Extensive experiments demonstrate the effectiveness, reliability and robustness of the proposed model.

AB - Semi-supervised multi-view learning is a remarkable but challenging task. Existing semi-supervised multi-view classification (SMVC) approaches mainly focus on performance improvement while ignoring decision reliability, which limits their deployment in safety-critical applications. Although several trusted multi-view classification methods are proposed recently, they rely on manual annotations. Therefore, this work emphasizes trusted multi-view classification learning under semi-supervised conditions. Different from existing SMVC methods, this work jointly models class probabilities and uncertainties based on evidential deep learning to formulate view-specific opinions. Moreover, unlike previous works that explore cross-view consistency in a single schema, this work proposes a multi-level consistency constraint. Specifically, we explore instance-level consistency on the view-specific representation space and category-level consistency on opinions from multiple views. Our proposed trusted graph-based contrastive loss nicely establishes the relationship between joint opinions and view-specific representations, which enables view-specific representations to enjoy a good manifold to improve classification performance. Overall, the proposed approach provides reliable and superior semi-supervised multi-view classification decisions. Extensive experiments demonstrate the effectiveness, reliability and robustness of the proposed model.

U2 - 10.1109/TMM.2024.3379079

DO - 10.1109/TMM.2024.3379079

M3 - Journal article

VL - 26

SP - 8268

EP - 8278

JO - IEEE Transactions on Multimedia

JF - IEEE Transactions on Multimedia

SN - 1520-9210

ER -