Comparer des méthodes

	Recherche d'architecture neuronale	Distillation de connaissances
Domaine	Apprentissage profond	Apprentissage profond
Famille	Machine learning	Machine learning
Année d'origine	2017	2015
Auteur d'origine	Zoph, B. & Le, Q.V.	Hinton, G., Vinyals, O. & Dean, J.
Type	Automated architecture optimization (deep learning)	Neural network compression (teacher–student)
Source fondatrice	Zoph, B. & Le, Q.V. (2017). Neural Architecture Search with Reinforcement Learning. ICLR.	Hinton, G., Vinyals, O. & Dean, J. (2015). Distilling the Knowledge in a Neural Network. NeurIPS Deep Learning Workshop.
Alias	Nöral Mimari Arama (NAS), NAS, automated architecture design, differentiable architecture search	Bilgi Damıtma (Knowledge Distillation), bilgi damıtma, teacher-student distillation, model distillation
Apparentées	5	5
Résumé	Neural Architecture Search (NAS), introduced by Zoph and Le in 2017, automatically optimizes architectural decisions such as a network's depth, width, and connection structure instead of hand-designing them. Leading methods in the field include DARTS, ENAS, and Once-for-All.	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.