EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Comparative Evaluation of Robust Loss Functions for Learning with Noisy Labels: From Synthetic to Real-World Annotations

Zhengchun Shang, Wenlan Wei

Journal of Computing Innovations and Applications · 17.03.2025 · 3 (1) · 57-65

DOI: 10.63575/cia.2025.30107 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computing Innovations and Applications, online erschienen am 17.03.2025. Verfasst von Zhengchun Shang und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Label noise is a pervasive challenge in supervised learning, as deep neural networks are prone to memorizing corrupted annotations during training. Robust loss functions have emerged as a principled approach to mitigate this issue without requiring auxiliary clean data or multi-network training pipelines. While numerous noise-tolerant and composite loss functions have been proposed in recent years, existing evaluations typically compare against only a limited subset of baselines under heterogeneous experimental protocols, making cross-method comparisons unreliable. This paper presents a systematic and unified comparative evaluation of eight representative robust loss functions—Cross-Entropy, Mean Absolute Error, Generalized Cross-Entropy, Symmetric Cross-Entropy, Active Passive Loss, Early-Learning Regularization, Generalized Jensen-Shannon divergence, and Active Negative Loss—across four distinct noise conditions: symmetric, asymmetric, instance-dependent, and real-world human annotation noise. All experiments are conducted on CIFAR-10 and CIFAR-100 with their corresponding CIFAR-N human-annotated noisy label sets (available at noisylabels.com), using a unified PreAct-ResNet-18 architecture and identical training protocol. The results reveal that composite and regularization-based losses consistently outperform simpler noise-tolerant alternatives, with the performance gap widening at elevated noise rates. Notably, method rankings observed under synthetic symmetric noise do not reliably transfer to real-world instance-dependent noise scenarios, highlighting the importance of evaluating under diverse noise conditions.

Zitieren

APA

Shang, Z., & Wei, W. (2025). A Comparative Evaluation of Robust Loss Functions for Learning with Noisy Labels: From Synthetic to Real-World Annotations. Journal of Computing Innovations and Applications, 3(1), 57-65. https://doi.org/10.63575/cia.2025.30107

Vancouver

Shang Z, Wei W. A Comparative Evaluation of Robust Loss Functions for Learning with Noisy Labels: From Synthetic to Real-World Annotations. Journal of Computing Innovations and Applications. 2025;3(1):57-65. doi:10.63575/cia.2025.30107

BibTeX
@article{Shang2025,
  author = {Shang, Zhengchun and Wei, Wenlan},
  title = {A Comparative Evaluation of Robust Loss Functions for Learning with Noisy Labels: From Synthetic to Real-World Annotations},
  journal = {Journal of Computing Innovations and Applications},
  year = {2025},
  volume = {3},
  number = {1},
  pages = {57--65},
  doi = {10.63575/cia.2025.30107}
}
Kostenlos starten →