Skip to content
DraconiticResearch
Search Sources API
203,203 studies31 sourcesUpdated 41 min ago
Peer-reviewedCC BY 4.0Mathematics

A Fuzzy Adjustment Mechanism to Tune Focal Loss Hyperparameters in Imbalanced Deep Learning

  • Tzung-Pei Hong1,2
  • Wei-Chun Peng2
  • Ja-Hwung Su1
  • Chun-Hao Chen3
  1. 1Department of Computer Science and Information Engineering, National University of Kaohsiung, Kaohsiung 811726, Taiwan
  2. 2Department of Computer Science and Engineering, National Sun Yat-sen University, Kaohsiung 804201, Taiwan
  3. 3Department of Computer Science and Information Engineering, National Kaohsiung University of Science and Technology, Kaohsiung 807618, Taiwan
Open at sourceDOICite

Abstract

To address the problem of data imbalance in deep learning, this paper proposes a fuzzy adjustment mechanism that dynamically adjusts the focal loss focusing parameter according to the class-specific learning status. Instead of using a single fixed focusing parameter for all classes, the proposed method adjusts the class-wise parameter based on three factors: class size, class-specific focal loss, and changes in focal loss. This enables the model to adjust the focus on different classes during training according to their learning conditions. The experimental results show that the proposed approach achieves a competitive performance compared with the evaluated fixed focal loss baselines. Based on paired t-tests across five trials, the proposed method shows statistically significant improvements over cross-entropy loss on both CIFAR-10 and SVHN datasets. For the fixed focal loss baselines, some differences are statistically significant, while others are not. Given the small number of paired trials (n = 5), these statistical results should be interpreted with caution. Overall, the proposed fuzzy dynamic mechanism provides a practical alternative to manually specifying a fixed focusing parameter by adaptively adjusting the parameter according to class-level learning conditions.

Related

  1. Bounded Adjustment with Reliability-Guided Embedding for Imbalanced Learning with Noisy Labels
    Mushir Akhtar, Akarsh J., M. Tanveer +1
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  2. On the Comparison of Optimizers for Imbalanced Learning
    Clément Lezane, François Bachoc, Jérôme Bolte +1
    HALPreprintCC BY 4.0Full textStatisticsHALarXiv
  3. ML-CLAPPS: A Probabilistic Framework for Imbalanced Multilabel Learning under Class Prioritization
    Priel Hazan, Nir Almog, Noam Keidar +2
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  4. Class-Conditioned Diffusion Augmentation and Explainable In-Context Ensemble Learning for Imbalanced Maintenance Action Classification in Aircraft Systems
    Begüm Erol
    MathematicsPeer-reviewedCC BY 4.0General mathematicsCrossref
  5. fSRD: Fuzzy Spectral Region Decomposition -- Automated Multi Operator Koopman Representations via an Adaptive Spectral Learning Architecture
    Charles Bokor, Mark Cary, Denise Morrey +1
    arXivPreprintCC BY 4.0Full textDynamical systemsarXiv

Cite

BibTeXRIS
BibTeX
@article{hong2026fuzzy,
  title = {{A Fuzzy Adjustment Mechanism to Tune Focal Loss Hyperparameters in Imbalanced Deep Learning}},
  author = {Tzung-Pei Hong and Wei-Chun Peng and Ja-Hwung Su and Chun-Hao Chen},
  journal = {Mathematics},
  year = {2026},
  month = oct,
  doi = {10.3390/math14213952},
  url = {https://www.mdpi.com/2227-7390/14/21/3952},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - A Fuzzy Adjustment Mechanism to Tune Focal Loss Hyperparameters in Imbalanced Deep Learning
AU  - Tzung-Pei Hong
AU  - Wei-Chun Peng
AU  - Ja-Hwung Su
AU  - Chun-Hao Chen
PY  - 2026
DA  - 2026/10/29
JO  - Mathematics
DO  - 10.3390/math14213952
UR  - https://www.mdpi.com/2227-7390/14/21/3952
AB  - To address the problem of data imbalance in deep learning, this paper proposes a fuzzy adjustment mechanism that dynamically adjusts the focal loss focusing parameter according to the class-specific learning status. Instead of using a single fixed focusing parameter for all classes, the proposed method adjusts the class-wise parameter based on three factors: class size, class-specific focal loss, and changes in focal loss. This enables the model to adjust the focus on different classes during training according to their learning conditions. The experimental results show that the proposed approach achieves a competitive performance compared with the evaluated fixed focal loss baselines. Based on paired t-tests across five trials, the proposed method shows statistically significant improvements over cross-entropy loss on both CIFAR-10 and SVHN datasets. For the fixed focal loss baselines, some differences are statistically significant, while others are not. Given the small number of paired trials (n = 5), these statistical results should be interpreted with caution. Overall, the proposed fuzzy dynamic mechanism provides a practical alternative to manually specifying a fixed focusing parameter by adaptively adjusting the parameter according to class-level learning conditions.
SN  - 2227-7390
ER  -