ScholarGate
Asistent

Uporedite metode

Pregledajte izabrane metode jednu pored druge; redovi koji se razlikuju su istaknuti.

Дистилација знања×Transferno učenje×
OblastDuboko učenjeMašinsko učenje
PorodicaMachine learningMachine learning
Godina nastanka20152010 (formalized); 1990s (early roots)
TvoracHinton, G., Vinyals, O. & Dean, J.Pan, S. J. & Yang, Q. (survey); Bengio, Y. (deep learning framing)
TipNeural network compression (teacher–student)Learning paradigm
Temeljni izvorHinton, G., Vinyals, O. & Dean, J. (2015). Distilling the Knowledge in a Neural Network. NeurIPS Deep Learning Workshop. link ↗Pan, S. J., & Yang, Q. (2010). A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 22(10), 1345–1359. DOI ↗
Drugi naziviBilgi Damıtma (Knowledge Distillation), bilgi damıtma, teacher-student distillation, model distillationTL, domain adaptation, fine-tuning, pre-trained model adaptation
Srodne53
SažetakKnowledge 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.Transfer learning is a machine learning paradigm in which knowledge gained from training a model on a source task or domain is reused to improve learning on a different but related target task or domain. It is especially powerful when labeled data for the target task is scarce, and it underlies most modern deep learning applications in computer vision, natural language processing, and beyond.
ScholarGateSkup podataka
  1. v1
  2. 2 Izvori
  3. PUBLISHED
  1. v1
  2. 2 Izvori
  3. PUBLISHED

Idi na pretragu Preuzmi slajdove

ScholarGateUporedite metode: Knowledge Distillation · Transfer Learning. Preuzeto 2026-06-16 sa https://scholargate.app/sr/compare