Mitigating Label Noise through Data Ambiguation

J. Lienen, E. Hüllermeier, ArXiv:2305.13764 (2023).

Preprint | English
Abstract
Label noise poses an important challenge in machine learning, especially in deep learning, in which large models with high expressive power dominate the field. Models of that kind are prone to memorizing incorrect labels, thereby harming generalization performance. Many methods have been proposed to address this problem, including robust loss functions and more complex label correction approaches. Robust loss functions are appealing due to their simplicity, but typically lack flexibility, while label correction usually adds substantial complexity to the training setup. In this paper, we suggest to address the shortcomings of both methodologies by "ambiguating" the target information, adding additional, complementary candidate labels in case the learner is not sufficiently convinced of the observed training label. More precisely, we leverage the framework of so-called superset learning to construct set-valued targets based on a confidence threshold, which deliver imprecise yet more reliable beliefs about the ground-truth, effectively helping the learner to suppress the memorization effect. In an extensive empirical evaluation, our method demonstrates favorable learning behavior on synthetic and real-world noise, confirming the effectiveness in detecting and correcting erroneous training labels.
Publishing Year
Journal Title
arXiv:2305.13764
LibreCat-ID

Cite this

Lienen J, Hüllermeier E. Mitigating Label Noise through Data Ambiguation. arXiv:230513764. Published online 2023.
Lienen, J., & Hüllermeier, E. (2023). Mitigating Label Noise through Data Ambiguation. In arXiv:2305.13764.
@article{Lienen_Hüllermeier_2023, title={Mitigating Label Noise through Data Ambiguation}, journal={arXiv:2305.13764}, author={Lienen, Julian and Hüllermeier, Eyke}, year={2023} }
Lienen, Julian, and Eyke Hüllermeier. “Mitigating Label Noise through Data Ambiguation.” ArXiv:2305.13764, 2023.
J. Lienen and E. Hüllermeier, “Mitigating Label Noise through Data Ambiguation,” arXiv:2305.13764. 2023.
Lienen, Julian, and Eyke Hüllermeier. “Mitigating Label Noise through Data Ambiguation.” ArXiv:2305.13764, 2023.
All files available under the following license(s):
Copyright Statement:
This Item is protected by copyright and/or related rights. [...]

Link(s) to Main File(s)
Access Level
Restricted Closed Access

Export

Marked Publications

Open Data LibreCat

Sources

arXiv 2305.13764

Search this title in

Google Scholar