Comparer des méthodes

	Apprentissage par transfert avec modélisation thématique	Classification basée sur BERT
Domaine	Apprentissage profond	Apprentissage profond
Famille	Machine learning	Machine learning
Année d'origine	2010s	2019
Auteur d'origine	Pan, S. J. & Yang, Q. (transfer learning survey); combined with Blei et al. (LDA, 2003)	Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (Google AI Language)
Type	Cross-domain adaptation of topic models	Pre-trained language model with fine-tuning
Source fondatrice	Pan, S. J., & Yang, Q. (2010). A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering, 22(10), 1345–1359.	Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of NAACL-HLT 2019 (pp. 4171–4186). Association for Computational Linguistics.
Alias	domain-transfer topic modeling, pretrained topic transfer, cross-domain topic adaptation, TL-LDA	BERT classifier, BERT fine-tuning for classification, BERT text classification, BERT-CLS
Apparentées	5	4
Résumé	Transfer Learning with Topic Modeling adapts topic structures discovered on a large or well-labeled source corpus to a related but distinct target domain where labeled data or large corpora are scarce. By reusing source-domain topic priors or pretrained embeddings as initialization, the approach produces richer, more coherent topics in the target domain than training from scratch.	BERT-based Classification fine-tunes Google's Bidirectional Encoder Representations from Transformers model on a labelled text dataset, replacing the generic pre-trained head with a task-specific classification layer. It exploits deep bidirectional context from hundreds of millions of pre-trained parameters to deliver state-of-the-art accuracy on short- and medium-length text classification tasks with relatively modest amounts of labelled data.