방법 비교

	자기 지도 학습	퓨샷 학습	전이 학습
분야	머신러닝	머신러닝	머신러닝
계열	Machine learning	Machine learning	Machine learning
기원 연도	2018–2020	2011–2017	2010 (formalized); 1990s (early roots)
창시자	LeCun, Y. and community (formalized ~2018–2020)	Lake, B. M.; Vinyals, O.; Finn, C. et al.	Pan, S. J. & Yang, Q. (survey); Bengio, Y. (deep learning framing)
유형	Representation learning paradigm	Meta-learning / low-data learning paradigm	Learning paradigm
원전	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/	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.	Pan, S. J., & Yang, Q. (2010). A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 22(10), 1345–1359.
별칭	SSL, self-supervised pre-training, pretext-task learning, unsupervised representation learning	FSL, low-shot learning, k-shot learning, meta-learning for few examples	TL, domain adaptation, fine-tuning, pre-trained model adaptation
관련	3	4	3
요약	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.	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.	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.