Comparer des méthodes

	Apprentissage Actif Robuste	Apprentissage semi-supervisé
Domaine	Apprentissage automatique	Apprentissage automatique
Famille	Machine learning	Machine learning
Année d'origine	2006	1970s–2006 (formalized)
Auteur d'origine	Balcan, M.-F.; Beygelzimer, A.; Langford, J.	Vapnik, V. N. and others (community of researchers, 1970s–2000s)
Type	Active learning with robustness guarantees	Learning paradigm
Source fondatrice	Balcan, M.-F., Beygelzimer, A., & Langford, J. (2006). Agnostic active learning. In Proceedings of the 23rd International Conference on Machine Learning (ICML 2006), pp. 65–72. ACM.	Chapelle, O., Scholkopf, B., & Zien, A. (Eds.) (2006). Semi-Supervised Learning. MIT Press. ISBN: 978-0-262-03358-9
Alias	RAL, noise-tolerant active learning, robust query learning, adversarially robust active learning	SSL, semi-supervised machine learning, transductive learning, label-efficient learning
Apparentées	6	5
Résumé	Robust Active Learning extends the standard active learning framework to handle noisy labels, adversarial perturbations, and unreliable oracles. Rather than assuming perfect labeling, it incorporates statistical or adversarial robustness guarantees into the query selection process, maintaining sample efficiency while tolerating corruption in the annotation process.	Semi-supervised learning (SSL) is a machine learning paradigm that trains models using a small set of labeled examples together with a much larger pool of unlabeled data. By leveraging the structure inherent in unlabeled data, SSL achieves accuracy closer to fully supervised models while requiring far fewer costly manual labels — making it practical when labeling is expensive, slow, or resource-constrained.