20.500.12556/DKUM-92602 Fostering fairness in image classification through awareness of sensitive data Machine learning (ML) has demonstrated remarkable ability to uncover hidden patterns in data. However, the presence of biases and discrimination originating from the data itself and, consequently, emerging in the ML outcomes, remains a pressing concern. With the exponential growth of unstructured data, such as images, fairness has become increasingly critical, as neural network (NN) models may inadvertently learn and perpetuate societal and historical biases. To address this challenge, we propose a fairness-aware loss function that iteratively prioritizes the worst-performing sensitive group during NN training. This approach aims to balance treatment quality across sensitive groups, achieving fairer image classification outcomes while incurring only a slight compromise in overall performance. Our method, evaluated on the FairFace dataset, demonstrates significant improvements in fairness metrics while maintaining comparable overall quality. These trade-offs highlight that the minor decrease in overall quality is justified by the improvement in fairness of the models. fairness search-basimage classification machine learning supervised learnign neural networks pravičnost strojno učenje nadzorovano učenje nevronske mreže true false true Elsevier Angleški jezik Slovenski jezik https://www.sciencedirect.com/journal/applied-soft-computing © 2025 The Authors Članek v reviji 2025-04-23 15:57:47 2025-04-23 15:57:47 2026-01-26 14:00:39 0000-00-00 00:00:00 2025 0 0 10 str. Vol. 136, [article] no.] 113090 May 2025 0000-00-00 Zaloznikova Objavljeno NiDoloceno 2024-11-05 2025-03-22 2025-04-04 004.8 232495363 10.1016/j.asoc.2025.113090 1872-9681 1-s2.0-S1568494625004016-main.pdf 1-s2.0-S1568494625004016-main.pdf 1 131792810C6ABE1A402EC5820F6B701B 3d6fe9d715383ca698d8180f612b823bd6e009b783736619d1498318b2bb1339 2f78121d-204b-11f0-adc7-00155d000105 https://dk.um.si/Dokument.php?lang=slv&id=190491 https://www.sciencedirect.com/science/article/pii/S1568494625004016?via%3Dihub#sec3 0 02e46636-fab7-11f0-9b4e-00155d00018d https://dk.um.si/Dokument.php?lang=slv&id=202561 Fakulteta za elektrotehniko, računalništvo in informatiko 0 0 0