Usporedite metode
Pregledajte odabrane metode jednu uz drugu; retci koji se razlikuju su istaknuti.
| Varijacijski automatski enkoder prilagodljiv domeni× | Prijenosno učenje× | |
|---|---|---|
| Područje≠ | Duboko učenje | Strojno učenje |
| Obitelj | Machine learning | Machine learning |
| Godina nastanka≠ | 2020 | 2010 (formalized); 1990s (early roots) |
| Tvorac≠ | Ilse, M.; Tomczak, J. M.; Louizos, C.; Welling, M. | Pan, S. J. & Yang, Q. (survey); Bengio, Y. (deep learning framing) |
| Vrsta≠ | Generative model with domain adaptation | Learning paradigm |
| Temeljni izvor≠ | Ilse, M., Tomczak, J. M., Louizos, C., & Welling, M. (2020). DIVA: Domain Invariant Variational Autoencoders. Proceedings of the Third Conference on Medical Imaging with Deep Learning (MIDL 2020), PMLR 121, 322–348. 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 nazivi | DA-VAE, domain-adaptive VAE, domain-conditioned variational autoencoder, cross-domain VAE | TL, domain adaptation, fine-tuning, pre-trained model adaptation |
| Srodne | 3 | 3 |
| Sažetak≠ | A Domain-Adaptive Variational Autoencoder (DA-VAE) extends the standard VAE framework to learn disentangled latent representations that separate domain-specific variation from class-relevant and domain-invariant content, enabling models trained on a source domain to generalise effectively to a different but related target domain with limited or no target labels. | 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 ↗ |
|
|