Hybrid dropout for deep ordinal classification
F. Bérchez-Moreno , F. Moreno-Cano, D. Guijo-Rubio, V. Vargas, P. Gutiérrez, C. Hervás-Martínez
18th International Work-Conference on Artificial Neural Networks, pp. 500-511, 2026Abstract
This paper presents a new application of a hybrid dropout technique for Ordinal Classification (OC), based on a novel regularisation method. Unlike standard dropout, which ignores class ordering, this hybrid dropout integrates ordinal information by adjusting neurons dropout probabilities based on their correlation with target labels. We evaluate its effectiveness using a ResNet18 architecture over three new OC datasets and compare it with the standard dropout approach and with an architecture with no dropout. Results show that the hybrid dropout consistently achieves the best performance across multiple well-known metrics (1-off, QWK, MAE, AMAE, and RPS), while also reducing prediction variability. Statistical analysis using the Wilcoxon signed-rank test confirms its robustness, obtaining 21 significant wins out of 30 comparisons, with no losses. These results highlight the importance of designing regularisation strategies that consider the problems ordinal structure, demonstrating that hybrid dropout effectively enhances generalisation and predictive accuracy.
Cite this publication
BibTex
@inproceedings{berchez-moreno2026hybrid,
author = {Francisco Bérchez-Moreno and Francisco Moreno-Cano and David Guijo-Rubio and Víctor Manuel Vargas and Pedro Antonio Gutiérrez and César Hervás-Martínez},
title = {Hybrid dropout for deep ordinal classification},
booktitle = {18th International Work-Conference on Artificial Neural Networks},
year = {2026},
pages = {500--511},
doi = {10.1007/978-3-032-02725-2_39}
}
APA
Bérchez-Moreno, F., Moreno-Cano, F., Guijo-Rubio, D., Vargas, V., Gutiérrez, P., Hervás-Martínez, C. (2026). Hybrid dropout for deep ordinal classification. In 18th International Work-Conference on Artificial Neural Networks (pp. 500-511).
CV
F. Bérchez-Moreno (CA), F. Moreno-Cano, D. Guijo-Rubio, V.M. Vargas, P.A. Gutiérrez, C. Hervás-Martínez, (4/6) "Hybrid dropout for deep ordinal classification". 18th International Work-Conference on Artificial Neural Networks, pp. 500-511, 2026.
RIS
TY - CONF T1 - Hybrid dropout for deep ordinal classification T2 - 18th International Work-Conference on Artificial Neural Networks AU - Bérchez-Moreno, Francisco AU - Moreno-Cano, Francisco AU - Guijo-Rubio, David AU - Vargas, Víctor Manuel AU - Gutiérrez, Pedro Antonio AU - Hervás-Martínez, César JO - 18th International Work-Conference on Artificial Neural Networks JA - 18th International Work-Conference on Artificial Neural Networks Y1 - 2026 PY - 2026 SP - 500 EP - 511 DO - 10.1007/978-3-032-02725-2_39 ER -