手法を比較

	転移学習	Few-shot Learning	自己教師あり学習
分野	機械学習	機械学習	機械学習
系統	Machine learning	Machine learning	Machine learning
提唱年	2010 (formalized); 1990s (early roots)	2011–2017	2018–2020
提唱者	Pan, S. J. & Yang, Q. (survey); Bengio, Y. (deep learning framing)	Lake, B. M.; Vinyals, O.; Finn, C. et al.	LeCun, Y. and community (formalized ~2018–2020)
種類	Learning paradigm	Meta-learning / low-data learning paradigm	Representation learning paradigm
原典	Pan, S. J., & Yang, Q. (2010). A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 22(10), 1345–1359.	Vinyals, O., Blundell, C., Lillicrap, T., Wierstra, D., & Kavukcuoglu, K. (2016). Matching Networks for One Shot Learning. Advances in Neural Information Processing Systems (NeurIPS), 29.	LeCun, Y. & Misra, I. (2022). Self-supervised learning: The dark matter of intelligence. Meta AI Blog. https://ai.facebook.com/blog/self-supervised-learning-the-dark-matter-of-intelligence/
別名	TL, domain adaptation, fine-tuning, pre-trained model adaptation	FSL, low-shot learning, k-shot learning, meta-learning for few examples	SSL, self-supervised pre-training, pretext-task learning, unsupervised representation learning
関連	3	4	3
概要	Transfer learning is a machine learning paradigm in which knowledge gained from training a model on a source task or domain is reused to improve learning on a different but related target task or domain. It is especially powerful when labeled data for the target task is scarce, and it underlies most modern deep learning applications in computer vision, natural language processing, and beyond.	Few-shot learning is a machine learning paradigm that trains models to recognize new classes or solve new tasks from only a handful of labeled examples — typically one to five — by leveraging prior knowledge acquired from a large, related training distribution. It is especially relevant in domains where labeling is expensive, scarce, or structurally limited.	Self-supervised learning (SSL) is a machine-learning paradigm that generates its own supervisory signal directly from unlabeled data by defining an auxiliary pretext task — such as predicting masked words, rotating images, or contrasting augmented views — and uses the learned representations as a powerful starting point for downstream tasks with minimal labeled examples.