Comparer des méthodes

	SVM univarié auto-supervisé	Apprentissage auto-supervisé
Domaine	Apprentissage automatique	Apprentissage automatique
Famille	Machine learning	Machine learning
Année d'origine	2018	2018–2020
Auteur d'origine	Golan & El-Yaniv; Ruff et al.	LeCun, Y. and community (formalized ~2018–2020)
Type	Self-supervised anomaly/novelty detection	Representation learning paradigm
Source fondatrice	Golan, I. & El-Yaniv, R. (2018). Deep One-Class Classification. Proceedings of the 35th International Conference on Machine Learning (ICML), PMLR 80, 1747–1756.	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	SS-OCSVM, Self-supervised SVDD, Self-supervised novelty detection, Pretext-task OC-SVM	SSL, self-supervised pre-training, pretext-task learning, unsupervised representation learning
Apparentées	6	3
Résumé	Self-supervised One-class SVM combines pretext-task-based representation learning with One-class SVM to detect anomalies and novelties without requiring labeled anomaly examples. The model first learns expressive feature embeddings from normal data alone, then fits an OC-SVM boundary in the learned feature space to flag out-of-distribution samples.	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.