Comparar métodos

	GANs de Aprendizagem por Transferência	Transfer Learning com Redes Neurais Convolucionais
Área	Aprendizado profundo	Aprendizado profundo
Família	Machine learning	Machine learning
Ano de origem	2014–2018	2010–2014
Autor original	Goodfellow, I. et al. (GAN); Wang & Ramanan (transfer to GAN)	Pan, S. J. & Yang, Q. (transfer learning framework); popularized for CNNs by Yosinski et al. and Razavian et al.
Tipo	Generative model with transferred weights	Transfer learning applied to convolutional neural networks
Fonte seminal	Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A. & Bengio, Y. (2014). Generative Adversarial Nets. Advances in Neural Information Processing Systems (NeurIPS), 27, 2672–2680.	Pan, S. J., & Yang, Q. (2010). A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 22(10), 1345–1359.
Outros nomes	TL-GAN, pretrained GAN, GAN fine-tuning, domain-adaptive GAN	TL-CNN, pretrained CNN, CNN fine-tuning, feature-extracting CNN
Relacionados	6	4
Resumo	Transfer Learning GAN initialises a Generative Adversarial Network — or both its generator and discriminator — from weights pretrained on a large source dataset, then fine-tunes the network on a smaller target dataset. This approach allows high-quality generative modelling even when target-domain data are scarce, by reusing low- and mid-level feature representations learned at scale.	Transfer Learning with CNN reuses a convolutional neural network that has already been trained on a large dataset — most commonly ImageNet — and adapts its learned feature detectors to a new, often smaller target dataset. This lets researchers achieve strong image-recognition performance without the massive compute and data resources required to train a CNN from scratch.