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。
| リカレントニューラルネットワークを用いた転移学習× | ファインチューニングされたリカレント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≠ | 2010 (TL survey); RNN: 1986 | 2015–2018 |
| 提唱者≠ | Pan, S. J. & Yang, Q. (transfer learning survey); RNN origins: Rumelhart, D. E. et al. (1986) | Popularised by Howard & Ruder (ULMFiT, 2018); RNN fine-tuning concept developed iteratively in the NLP community from ~2015 |
| 種類≠ | Transfer learning on sequence model | Transfer learning / sequential model adaptation |
| 原典≠ | Pan, S. J., & Yang, Q. (2010). A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 22(10), 1345–1359. DOI ↗ | Howard, J. & Ruder, S. (2018). Universal Language Model Fine-Tuning for Text Classification. Proceedings of ACL 2018, 328–339. DOI ↗ |
| 別名 | TL-RNN, Pretrained RNN, RNN Transfer Learning, Recurrent Transfer Learning | Fine-Tuned RNN, RNN Fine-Tuning, domain-adapted RNN, pre-trained RNN with downstream adaptation |
| 関連≠ | 5 | 6 |
| 概要≠ | Transfer Learning with Recurrent Neural Network (TL-RNN) reuses weights learned by an RNN on a large source task — such as language modelling or sequence prediction — and adapts them to a new, often smaller target task. This strategy lets practitioners obtain strong sequence-modelling performance without the need for massive labelled datasets. | A Fine-Tuned Recurrent Neural Network (RNN) starts from a model pre-trained on large corpora or time-series data and adapts its weights to a specific downstream task through controlled gradient updates. The approach dramatically cuts the labeled data needed for strong sequence modeling performance in text classification, named entity recognition, sentiment analysis, and related tasks. |
| ScholarGateデータセット ↗ |
|
|