Bridging the gap in FER: addressing age bias in deep learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gaya-Morey, F. Xavier, Sanchez-Perez, Julia, Manresa-Yee, Cristina, Buades-Rubio, Jose M.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915381868756992
author Gaya-Morey, F. Xavier
Sanchez-Perez, Julia
Manresa-Yee, Cristina
Buades-Rubio, Jose M.
author_facet Gaya-Morey, F. Xavier
Sanchez-Perez, Julia
Manresa-Yee, Cristina
Buades-Rubio, Jose M.
contents Facial Expression Recognition (FER) systems based on deep learning have achieved impressive performance in recent years. However, these models often exhibit demographic biases, particularly with respect to age, which can compromise their fairness and reliability. In this work, we present a comprehensive study of age-related bias in deep FER models, with a particular focus on the elderly population. We first investigate whether recognition performance varies across age groups, which expressions are most affected, and whether model attention differs depending on age. Using Explainable AI (XAI) techniques, we identify systematic disparities in expression recognition and attention patterns, especially for "neutral", "sadness", and "anger" in elderly individuals. Based on these findings, we propose and evaluate three bias mitigation strategies: Multi-task Learning, Multi-modal Input, and Age-weighted Loss. Our models are trained on a large-scale dataset, AffectNet, with automatically estimated age labels and validated on balanced benchmark datasets that include underrepresented age groups. Results show consistent improvements in recognition accuracy for elderly individuals, particularly for the most error-prone expressions. Saliency heatmap analysis reveals that models trained with age-aware strategies attend to more relevant facial regions for each age group, helping to explain the observed improvements. These findings suggest that age-related bias in FER can be effectively mitigated using simple training modifications, and that even approximate demographic labels can be valuable for promoting fairness in large-scale affective computing systems.
format Preprint
id arxiv_https___arxiv_org_abs_2507_07638
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bridging the gap in FER: addressing age bias in deep learning
Gaya-Morey, F. Xavier
Sanchez-Perez, Julia
Manresa-Yee, Cristina
Buades-Rubio, Jose M.
Computer Vision and Pattern Recognition
Facial Expression Recognition (FER) systems based on deep learning have achieved impressive performance in recent years. However, these models often exhibit demographic biases, particularly with respect to age, which can compromise their fairness and reliability. In this work, we present a comprehensive study of age-related bias in deep FER models, with a particular focus on the elderly population. We first investigate whether recognition performance varies across age groups, which expressions are most affected, and whether model attention differs depending on age. Using Explainable AI (XAI) techniques, we identify systematic disparities in expression recognition and attention patterns, especially for "neutral", "sadness", and "anger" in elderly individuals. Based on these findings, we propose and evaluate three bias mitigation strategies: Multi-task Learning, Multi-modal Input, and Age-weighted Loss. Our models are trained on a large-scale dataset, AffectNet, with automatically estimated age labels and validated on balanced benchmark datasets that include underrepresented age groups. Results show consistent improvements in recognition accuracy for elderly individuals, particularly for the most error-prone expressions. Saliency heatmap analysis reveals that models trained with age-aware strategies attend to more relevant facial regions for each age group, helping to explain the observed improvements. These findings suggest that age-related bias in FER can be effectively mitigated using simple training modifications, and that even approximate demographic labels can be valuable for promoting fairness in large-scale affective computing systems.
title Bridging the gap in FER: addressing age bias in deep learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.07638