Sammenlign metoder
Gennemgå dine valgte metoder side om side; rækker, der afviger, er fremhævet.
| Semi-supervised Metric Learning× | Selvovervåget læring× | |
|---|---|---|
| Fagområde | Maskinlæring | Maskinlæring |
| Familie | Machine learning | Machine learning |
| Oprindelsesår≠ | 2007–2008 | 2018–2020 |
| Ophavsperson≠ | Yeung, D.-Y. & Chang, H.; Davis, J. V. & Dhillon, I. S. | LeCun, Y. and community (formalized ~2018–2020) |
| Type≠ | Hybrid supervised/unsupervised distance learning | Representation learning paradigm |
| Oprindelig kilde≠ | Yeung, D.-Y., & Chang, H. (2007). A kernel approach for semi-supervised metric learning. IEEE Transactions on Neural Networks, 18(1), 141–149. DOI ↗ | 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/ link ↗ |
| Aliasser | SSML, semi-supervised distance learning, constrained metric learning, weakly supervised metric learning | SSL, self-supervised pre-training, pretext-task learning, unsupervised representation learning |
| Relaterede≠ | 5 | 3 |
| Resumé≠ | 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. |
| ScholarGateDatasæt ↗ |
|
|