Evidential Generative Adversarial Networks for Handling Imbalanced Learning - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Evidential Generative Adversarial Networks for Handling Imbalanced Learning

Résumé

The predictive performance of machine learning models tends to deteriorate in the presence of class imbalance. Multiple strategies have been proposed to address this issue. A popular strategy consists of oversampling the minority class. Classic approaches such as SMOTE utilize techniques like nearest neighbor search and linear interpolation, which can pose difficulties when dealing with datasets that have a large number of dimensions and intricate data distributions. As a way to create synthetic examples in the minority class, Generative Adversarial Networks (GANs) have been suggested as an alternative technique due to their ability to simulate complex data distributions. However, most GAN-based oversampling methods tend to ignore data uncertainty. In this paper, we propose a novel GAN-based oversampling method using evidence theory. An auxiliary evidential classifier is incorporated in the GAN architecture in order to guide the training process of the generative model. The objective is to push GAN to generate minority objects at the borderline of the minority class, near difficult-to-classify objects. Through extensive analysis, we demonstrate that the proposed approach provides better performance, compared to other popular methods.
ECSQARU23_grina_final.pdf (341.1 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Fichier non déposé

Dates et versions

hal-04432825 , version 1 (01-02-2024)

Identifiants

Citer

Fares Grina, Zied Elouedi, Eric Lefevre. Evidential Generative Adversarial Networks for Handling Imbalanced Learning. 17th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, ECSQARU'2023, Sep 2023, Arras, France. pp.257-269, ⟨10.1007/978-3-031-45608-4_20⟩. ⟨hal-04432825⟩

Collections

UNIV-ARTOIS LGI2A
11 Consultations
4 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More