Comparer des méthodes

	Apprentissage semi-supervisé	Apprentissage auto-supervisé
Domaine	Apprentissage automatique	Apprentissage automatique
Famille	Machine learning	Machine learning
Année d'origine	1970s–2006 (formalized)	2018–2020
Auteur d'origine	Vapnik, V. N. and others (community of researchers, 1970s–2000s)	LeCun, Y. and community (formalized ~2018–2020)
Type	Learning paradigm	Representation learning paradigm
Source fondatrice	Chapelle, O., Scholkopf, B., & Zien, A. (Eds.) (2006). Semi-Supervised Learning. MIT Press. ISBN: 978-0-262-03358-9	LeCun, Y. & Misra, I. (2022). Self-supervised learning: The dark matter of intelligence. Meta AI Blog. https://ai.facebook.com/blog/self-supervised-learning-the-dark-matter-of-intelligence/
Alias	SSL, semi-supervised machine learning, transductive learning, label-efficient learning	SSL, self-supervised pre-training, pretext-task learning, unsupervised representation learning
Apparentées	5	3
Résumé	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.	Self-supervised learning (SSL) is a machine-learning paradigm that generates its own supervisory signal directly from unlabeled data by defining an auxiliary pretext task — such as predicting masked words, rotating images, or contrasting augmented views — and uses the learned representations as a powerful starting point for downstream tasks with minimal labeled examples.