IRNet: Iterative Refinement Network for Noisy Partial Label Learning

Partial label learning (PLL) is a typical weakly supervised learning, where each sample is associated with a set of candidate labels. Its basic assumption is that the ground-truth label must be in the candidate set, but this assumption may not be satisfied due to the unprofessional judgment of annot...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on pattern analysis and machine intelligence Jg. PP; S. 1 - 17
Hauptverfasser: Lian, Zheng, Xu, Mingyu, Chen, Lan, Sun, Licai, Liu, Bin, Feng, Lei, Tao, Jianhua
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States IEEE 13.10.2025
Schlagworte:
Annotations
Electronic mail
Entropy
Faces
Iterative methods
Iterative Refinement Network (IRNet)
label correction
Maximum likelihood estimation
multi-round refinement
Noise measurement
Noise robustness
noisy partial label learning
noisy sample detection
Phase locked loops
Training
ISSN:0162-8828, 1939-3539, 2160-9292, 1939-3539
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Partial label learning (PLL) is a typical weakly supervised learning, where each sample is associated with a set of candidate labels. Its basic assumption is that the ground-truth label must be in the candidate set, but this assumption may not be satisfied due to the unprofessional judgment of annotators. Therefore, we relax this assumption and focus on a more general task, noisy PLL, where the ground-truth label may not exist in the candidate set. To address this challenging task, we propose a novel framework called "Iterative Refinement Network (IRNet)", aiming to purify noisy samples through two key modules (i.e., noisy sample detection and label correction). To achieve better performance, we exploit smoothness constraints to reduce prediction errors in these modules. Through theoretical analysis, we prove that IRNet is able to reduce the noise level of the dataset and eventually approximate the Bayes optimal classifier. Meanwhile, IRNet is a plug-in strategy that can be integrated with existing PLL approaches. Experimental results on multiple benchmark datasets show that IRNet outperforms state-of-the-art approaches on noisy PLL. Our source code is available at: https://github.com/zeroQiaoba/IRNet .
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0162-8828
1939-3539
2160-9292
1939-3539
DOI:10.1109/TPAMI.2025.3620388