Mining bias-target Alignment from Voronoi Cells - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Mining bias-target Alignment from Voronoi Cells

Résumé

Despite significant research efforts, deep neural networks are still vulnerable to biases: this raises concerns about their fairness and limits their generalization. In this paper, we propose a bias-agnostic approach to mitigate the impact of bias in deep neural networks. Unlike traditional debiasing approaches, we rely on a metric to quantify ``bias alignment/misalignment'' on target classes, and use this information to discourage the propagation of bias-target alignment information through the network. We conduct experiments on several commonly used datasets for debiasing and compare our method to supervised and bias-specific approaches. Our results indicate that the proposed method achieves comparable performance to state-of-the-art supervised approaches, although it is bias-agnostic, even in presence of multiple biases in the same sample.

Dates et versions

hal-04347752 , version 1 (15-12-2023)

Identifiants

Citer

Rémi Nahon, Van-Tam Nguyen, Enzo Tartaglione. Mining bias-target Alignment from Voronoi Cells. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Oct 2023, Paris, France. ⟨hal-04347752⟩
15 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More