השוואת שיטות

	משתמש אוטומטי וריאציוני למחצה-מפוקח	למידת העברה עם מקודד אוטומטי וריאציוני
תחום	למידה עמוקה	למידה עמוקה
משפחה	Machine learning	Machine learning
שנת המקור	2014	2014 (VAE); 2010 (transfer learning survey)
הוגה השיטה	Kingma, D. P.; Mohamed, S.; Rezende, D. J.; Wierstra, D.	Kingma, D. P. & Welling, M. (VAE); transfer learning framework from Pan & Yang
סוג	Generative probabilistic model (semi-supervised)	Generative model with transferred encoder/decoder
מקור מכונן	Kingma, D. P., Mohamed, S., Rezende, D. J., & Wierstra, D. (2014). Semi-supervised learning with deep generative models. Advances in Neural Information Processing Systems (NeurIPS), 27, 3581–3589.	Kingma, D. P., & Welling, M. (2014). Auto-Encoding Variational Bayes. International Conference on Learning Representations (ICLR 2014).
כינויים	Semi-supervised VAE, M2 model, VAE with label propagation, deep generative semi-supervised model	TL-VAE, pretrained VAE, VAE transfer learning, fine-tuned variational autoencoder
קשורות	6	6
תקציר	The semi-supervised VAE (M2 model) is a deep generative method that jointly learns a latent representation of inputs and a classifier, leveraging both labeled and unlabeled examples in a principled probabilistic framework. Introduced by Kingma et al. in 2014, it allows accurate classification even when labels are scarce by having the generative model explain away unlabeled observations.	Transfer Learning with a Variational Autoencoder (TL-VAE) reuses an encoder and/or decoder pre-trained on a large source dataset and adapts it to a smaller target domain. By inheriting a rich probabilistic latent space rather than starting from random weights, TL-VAE dramatically reduces the amount of target-domain data needed for high-quality generation or representation learning.