Linganisha mbinu

	Jifunze kipimo cha kukabiliana na hali kwa kutumia nusu-usimamizi	Jifunze kwa Kujisimamia
Nyanja	Ujifunzaji wa Mashine	Ujifunzaji wa Mashine
Familia	Machine learning	Machine learning
Mwaka wa asili	2007–2008	2018–2020
Mwanzilishi	Yeung, D.-Y. & Chang, H.; Davis, J. V. & Dhillon, I. S.	LeCun, Y. and community (formalized ~2018–2020)
Aina	Hybrid supervised/unsupervised distance learning	Representation learning paradigm
Chanzo asilia	Yeung, D.-Y., & Chang, H. (2007). A kernel approach for semi-supervised metric learning. IEEE Transactions on Neural Networks, 18(1), 141–149.	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/
Majina mbadala	SSML, semi-supervised distance learning, constrained metric learning, weakly supervised metric learning	SSL, self-supervised pre-training, pretext-task learning, unsupervised representation learning
Zinazohusiana	5	3
Muhtasari	Semi-supervised metric learning learns a task-adapted distance function by combining a small set of labeled pairwise constraints — must-link and cannot-link pairs — with the geometric structure of a much larger pool of unlabeled data. The result is a Mahalanobis-style or kernel-based distance that reflects both supervision and data topology, improving downstream tasks such as nearest-neighbor classification and clustering.	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.