OpenBPR: Bias-Guided Pseudo-Label Refinement for Open-World Semi-Supervised Learning
IEEE Transactions on Circuits and Systems for Video Technology · 2026-08 · 36 (8) · 11858-11875
DOI: 10.1109/tcsvt.2026.3685347 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2026-08. Verfasst von Guanjia Zhang und 5 weiteren Autor:innen. Die Arbeit nennt 67 Literaturangaben. Gefördert von: Beijing Natural Science Foundation.
Abstract (Originalsprache)
Semi-supervised learning (SSL) enhances model generalizability by jointly leveraging labeled and unlabeled data. Nevertheless, the closed-world assumption of SSL always fails in open-world scenarios, where unlabeled data often contains novel classes. To address this limitation, open-world SSL (OWSSL) has been proposed as a more realistic paradigm, aiming not only to recognize known classes but also to discover novel classes. Existing OWSSL methods typically rely on representation similarity and pseudo-labeling to discriminate classes. However, during model training, these methods neglect the inherent class-prediction bias, consequently leading to self-reinforcing confirmation bias in pseudo-labels and representation confusion for hard novel classes. To address these critical challenges, we propose a Open-world Bias-guided Pseudo-label Refinement approach, named OpenBPR, which is the first to regard class prediction bias as the reference to guide the debiased pseudo-labeling and class representation decoupling. In OpenBPR, we propose a debiased pseudo-labeling method based on expectation-maximization, which exploits class prediction bias to dynamically optimize pseudo-labels, effectively alleviating confirmation bias in pseudo-labels. Furthermore, we propose a class-aware representation decoupling strategy for hard novel classes, which decouples representations by the designed competitive class decoupling regularization to assist in improving the refinement performance of pseudo-labels. Experimental results on a series of benchmark datasets demonstrate that OpenBPR outperforms state-of-the-art methods in discriminating both known and novel classes.
Förderung
- Beijing Natural Science Foundation
Zitieren
APA
Zhang, G., Xing, W., Liu, W., Sang, F., Wang, X., & Wei, X. (2026). OpenBPR: Bias-Guided Pseudo-Label Refinement for Open-World Semi-Supervised Learning. IEEE Transactions on Circuits and Systems for Video Technology, 36(8), 11858-11875. https://doi.org/10.1109/tcsvt.2026.3685347
Vancouver
Zhang G, Xing W, Liu W, Sang F, Wang X, Wei X. OpenBPR: Bias-Guided Pseudo-Label Refinement for Open-World Semi-Supervised Learning. IEEE Transactions on Circuits and Systems for Video Technology. 2026;36(8):11858-11875. doi:10.1109/tcsvt.2026.3685347
BibTeX
@article{Zhang2026,
author = {Zhang, Guanjia and Xing, Weiwei and Liu, Weibin and Sang, Fusong and Wang, Xinyue and Wei, Xiang},
title = {OpenBPR: Bias-Guided Pseudo-Label Refinement for Open-World Semi-Supervised Learning},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
year = {2026},
volume = {36},
number = {8},
pages = {11858--11875},
doi = {10.1109/tcsvt.2026.3685347}
}
EUVIMED