Soft labelling for deep ordinal classification: an experimental review
Víctor Manuel Vargas-Yun 
David Guijo-Rubio 
Rafael Ayllón-Gavilán 
Antonio Manuel Gómez-Orellana 
Pedro Antonio Gutiérrez 
César Hervás-Martínez IEEE Transactions on Knowledge and Data Engineering
COMPUTER SCIENCE, INFORMATION SYSTEMS
Abstract
Ordinal classification, where labels follow a natural order, has gained increasing attention, particularly in the deep learning community due to its relevance in tasks such as age estimation, medical grading, and quality assessment. Despite the growing number of deep ordinal classification methods, a comprehensive experimental analysis of their core ordinal components remains lacking. This work presents a systematic evaluation of deep ordinal classifiers by analysing the impact of three key modelling choices: the loss function, output layer, and labelling strategy. To analyse their effects, we adopt a unified architecture and evaluate one nominal and 19 ordinal configurations, resulting from combination of two loss functions, two output layers, and five labelling strategies. These configurations are assessed on 12 diverse ordinal image datasets using six performance metrics, including both ordinal and nominal measures. Results show that ordinal output layers consistently outperform softmax, and that soft labelling generally improves generalisation. While categorical cross-entropy achieves better average performance, especially on nominal metrics, no configuration performs best across all datasets. Statistical analyses indicate significant interactions between losses, outputs, labelling strategies, and datasets, highlighting the need to adapt methodological choices to specific tasks. These findings provide valuable guidance for designing robust deep ordinal classification models.
BibTex Citation
@article{Vargas2026Soft,
author = {Vargas-Yun, V{\' i}ctor Manuel and Guijo-Rubio, David and Ayll{\' o}n-Gavil{\' a}n, Rafael and G{\' o}mez-Orellana, Antonio Manuel and Guti{\' e}rrez, Pedro Antonio and Herv{\' a}s-Mart{\' i}nez, C{\' e}sar},
journal = {IEEE Transactions on Knowledge and Data Engineering},
doi = {10.1109/TKDE.2026.3681678},
year = {2026},
pages = {1--20},
publisher = {IEEE},
title = {Soft labelling for deep ordinal classification: an experimental review},
}
BibTex Unicode Citation
@article{Vargas2026Soft,
author = {Vargas-Yun, Víctor Manuel and Guijo-Rubio, David and Ayllón-Gavilán, Rafael and Gómez-Orellana, Antonio Manuel and Gutiérrez, Pedro Antonio and Hervás-Martínez, César},
journal = {IEEE Transactions on Knowledge and Data Engineering},
doi = {10.1109/TKDE.2026.3681678},
year = {2026},
pages = {1--20},
publisher = {IEEE},
title = {Soft labelling for deep ordinal classification: an experimental review},
}
APA Citation
Vargas-Yun, V. M., Guijo-Rubio, D., Ayllón-Gavilán, R., Gómez-Orellana, A. M., Gutiérrez, P. A., & Hervás-Martínez, C. (2026). Soft labelling for deep ordinal classification: an experimental review. IEEE Transactions on Knowledge and Data Engineering, 1–20. https://doi.org/10.1109/TKDE.2026.3681678
RIS Citation
TY - JOUR
AU - Vargas-Yun, Víctor Manuel
AU - Guijo-Rubio, David
AU - Ayllón-Gavilán, Rafael
AU - Gómez-Orellana, Antonio Manuel
AU - Gutiérrez, Pedro Antonio
AU - Hervás-Martínez, César
DA - 2026///
PY - 2026
DO - 10.1109/TKDE.2026.3681678
ID - temp_id_193398719462
SP - 1-20
T2 - IEEE Transactions on Knowledge and Data Engineering
TI - Soft labelling for deep ordinal classification: an experimental review
ER -
CV Citation
V.M. Vargas-Yun, D. Guijo-Rubio (CA), R. Ayllón-Gavilán, A.M. Gómez-Orellana, P.A. Gutiérrez, C. Hervás-Martínez (1/6). "Soft labelling for deep ordinal classification: an experimental review". IEEE Transactions on Knowledge and Data Engineering, pp. 1–20, 2026. (Q1D1, IF: 10.4).