Comparer des méthodes

	Perceptron multicouche faiblement supervisé	Perceptron multicouche affiné
Domaine	Apprentissage profond	Apprentissage profond
Famille	Machine learning	Machine learning
Année d'origine	2016–2018	1986 (MLP); fine-tuning practice formalised c. 2014
Auteur d'origine	Multiple contributors; paradigm formalized by Zhou (2018) and Ratner et al. (2016)	Rumelhart, Hinton & Williams (MLP); Yosinski et al. (fine-tuning analysis)
Type	Feedforward neural network trained under weak supervision	Supervised deep learning with pre-trained weight initialisation
Source fondatrice	Zhou, Z.-H. (2018). A brief introduction to weakly supervised learning. National Science Review, 5(1), 44–53.	Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323, 533–536.
Alias	WS-MLP, weakly supervised feedforward network, noisy-label MLP, weak-label multilayer perceptron	fine-tuned MLP, adapted MLP, domain-adapted multilayer perceptron, MLP fine-tuning
Apparentées	5	4
Résumé	A Weakly Supervised Multilayer Perceptron trains a standard feedforward neural network when only imperfect supervision is available — labels may be noisy, incomplete, crowd-sourced, rule-generated, or derived from distant supervision — enabling learning at scale without the cost of full expert annotation.	A Fine-Tuned Multilayer Perceptron starts from weights learned on a source task — or a large general-purpose dataset — and continues training on a smaller target dataset with a reduced learning rate. This reuse of pre-learned representations allows the MLP to converge faster and generalise better than training from scratch, especially when labelled target data is scarce.