Compara mètodes
Revisa els mètodes seleccionats l'un al costat de l'altre; les files que difereixen es ressalten.
| Perceptró Multicapa Finament Ajustat× | Perceptró Multicapa (MLP)× | |
|---|---|---|
| Camp | Aprenentatge profund | Aprenentatge profund |
| Família | Machine learning | Machine learning |
| Any d'origen≠ | 1986 (MLP); fine-tuning practice formalised c. 2014 | 1986 |
| Autor original≠ | Rumelhart, Hinton & Williams (MLP); Yosinski et al. (fine-tuning analysis) | Rumelhart, D. E.; Hinton, G. E.; Williams, R. J. |
| Tipus≠ | Supervised deep learning with pre-trained weight initialisation | Supervised feedforward neural network |
| Font seminal≠ | Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323, 533–536. DOI ↗ | Rumelhart, D. E., Hinton, G. E. & Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323, 533–536. DOI ↗ |
| Àlies≠ | fine-tuned MLP, adapted MLP, domain-adapted multilayer perceptron, MLP fine-tuning | MLP, feedforward neural network, fully connected neural network, vanilla neural network |
| Relacionats | 4 | 4 |
| Resum≠ | A Fine-Tuned Multilayer Perceptron starts from weights learned on a source task — or a large general-purpose dataset — and continues training on a smaller target dataset with a reduced learning rate. This reuse of pre-learned representations allows the MLP to converge faster and generalise better than training from scratch, especially when labelled target data is scarce. | A Multilayer Perceptron is a classic fully connected feedforward neural network trained with the backpropagation algorithm, as formalised by Rumelhart, Hinton & Williams in their landmark 1986 Nature paper. Composed of an input layer, one or more hidden layers of neurons, and an output layer, the MLP learns nonlinear mappings from input features to target outputs and serves as the foundational building block of modern deep learning. |
| ScholarGateConjunt de dades ↗ |
|
|