Neural Networks Bias Mitigation Through Fuzzy Logic and Saliency Maps

Sahar Shah, Davide Ciucci, Sara Manzoni, Italo Zoppis

2025

Abstract

Mitigating biases in neural networks is crucial to reduce or eliminate the predictive model’s unfair responses, which may arise from unbalanced training, defective architectures, or even social prejudices embedded in the data. This study proposes a novel and fully differentiable framework for mitigating neural network bias using Saliency Maps and Fuzzy Logic. We focus our analysis on a simulation study for recommendation systems, where neural networks are crucial in classifying job applicants based on relevant and sensitive attributes. Leveraging the interpretability of a set of Fuzzy implications and the importance of features attributed by Saliency Maps, our approach penalizes models when they overly rely on biased predictions during training. In this way, we ensure that bias mitigation occurs within the gradient-based optimization process, allowing efficient model training and evaluation.

Paper Citation

in Harvard Style

Shah S., Ciucci D., Manzoni S. and Zoppis I. (2025). Neural Networks Bias Mitigation Through Fuzzy Logic and Saliency Maps. In Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART; ISBN 978-989-758-737-5, SciTePress, pages 1343-1351. DOI: 10.5220/0013366800003890

in Bibtex Style

@conference{icaart25,
author={Sahar Shah and Davide Ciucci and Sara Manzoni and Italo Zoppis},
title={Neural Networks Bias Mitigation Through Fuzzy Logic and Saliency Maps},
booktitle={Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART},
year={2025},
pages={1343-1351},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013366800003890},
isbn={978-989-758-737-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART
TI - Neural Networks Bias Mitigation Through Fuzzy Logic and Saliency Maps
SN - 978-989-758-737-5
AU - Shah S.
AU - Ciucci D.
AU - Manzoni S.
AU - Zoppis I.
PY - 2025
SP - 1343
EP - 1351
DO - 10.5220/0013366800003890
PB - SciTePress