Comparer des méthodes

	Confidentialité différentielle	Distillation de connaissances
Domaine	Protection de la vie privée	Apprentissage profond
Famille	Machine learning	Machine learning
Année d'origine	2006	2015
Auteur d'origine	Cynthia Dwork	Hinton, G., Vinyals, O. & Dean, J.
Type	Privacy-preserving randomized mechanism	Neural network compression (teacher–student)
Source fondatrice	Dwork, C. (2006). Differential privacy. International Colloquium on Automata, Languages and Programming (ICALP), 1–12.	Hinton, G., Vinyals, O. & Dean, J. (2015). Distilling the Knowledge in a Neural Network. NeurIPS Deep Learning Workshop.
Alias	DP, epsilon-differential privacy, randomized privacy, Diferansiyel Gizlilik	Bilgi Damıtma (Knowledge Distillation), bilgi damıtma, teacher-student distillation, model distillation
Apparentées	3	5
Résumé	Differential privacy is a mathematical framework for releasing statistical information about a dataset while providing rigorous guarantees that individual records cannot be identified or inferred. Introduced by Cynthia Dwork in 2006, it formalizes privacy as a probabilistic bound: any single individual's presence or absence in the dataset changes the output distribution by at most a multiplicative factor of e^ε, where ε is the privacy budget controlling the privacy–utility tradeoff.	Knowledge Distillation is a model-compression technique, introduced by Geoffrey Hinton and colleagues in 2015, that trains a small student model using the soft-label outputs of a large teacher model. Distilled models such as DistilBERT and TinyBERT reach roughly 97% of the larger model's performance while running far faster.