手法を比較
選択した手法を並べて確認できます。異なる行はハイライト表示されます。
| ファイン・チューニングされた多層パーセプトロン× | ファインチューニングされた畳み込みニューラルネットワーク× | |
|---|---|---|
| 分野 | 深層学習 | 深層学習 |
| 系統 | Machine learning | Machine learning |
| 提唱年≠ | 1986 (MLP); fine-tuning practice formalised c. 2014 | 2012–2014 |
| 提唱者≠ | Rumelhart, Hinton & Williams (MLP); Yosinski et al. (fine-tuning analysis) | Yosinski, J. et al. (theoretical basis); practice widespread from Krizhevsky et al. 2012 onward |
| 種類≠ | Supervised deep learning with pre-trained weight initialisation | Transfer learning technique (supervised fine-tuning) |
| 原典≠ | Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323, 533–536. DOI ↗ | Yosinski, J., Clune, J., Bengio, Y., & Lipson, H. (2014). How transferable are features in deep neural networks? Advances in Neural Information Processing Systems, 27. link ↗ |
| 別名 | fine-tuned MLP, adapted MLP, domain-adapted multilayer perceptron, MLP fine-tuning | Fine-tuned CNN, CNN fine-tuning, CNN transfer learning with fine-tuning, adapted convolutional network |
| 関連≠ | 4 | 5 |
| 概要≠ | 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. | Fine-tuning a CNN means starting from a network already trained on a large dataset — typically ImageNet — and continuing training on a smaller target dataset so the model adapts its learned visual features to a new task. This approach dramatically reduces the data and compute required to reach strong performance compared with training from scratch. |
| ScholarGateデータセット ↗ |
|
|