Active learning algorithm through the lens of rejection arguments - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2023

Active learning algorithm through the lens of rejection arguments

Résumé

Active learning is a paradigm of machine learning which aims at reducing the amount of labeled data needed to train a classifier. Its overall principle is to sequentially select the most informative data points, which amounts to determining the uncertainty of regions of the input space. The main challenge lies in building a procedure that is computationally efficient and that offers appealing theoretical properties; most of the current methods satisfy only one or the other. In this paper, we use the classification with rejection in a novel way to estimate the uncertain regions. We provide an active learning algorithm and prove its theoretical benefits under classical assumptions. In addition to the theoretical results, numerical experiments have been carried out on synthetic and non-synthetic datasets. These experiments provide empirical evidence that the use of rejection arguments in our active learning algorithm is beneficial and allows good performance in various statistical situations.
Fichier principal
Vignette du fichier
activeLearningReject.pdf (1.86 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03764630 , version 1 (30-08-2022)

Identifiants

Citer

Christophe Denis, Mohamed Hebiri, Boris Ndjia Njike, Xavier Siebert. Active learning algorithm through the lens of rejection arguments. 2022. ⟨hal-03764630⟩
17 Consultations
30 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More