Krahasoni metodat

	Mësimi Online	Rrjeti Neural Siamese
Fusha	Mësimi i makinës	Mësimi i thellë
Familja	Machine learning	Machine learning
Viti i origjinës	1958–2000s	1993
Krijuesi	Rosenblatt, F.; Littlestone, N.; Shalev-Shwartz, S. (key contributors)	Jane Bromley & Yann LeCun et al.; popularized by Koch et al.
Lloji	Learning paradigm (sequential model update)	Deep metric-learning architecture
Burimi themelues	Shalev-Shwartz, S. (2011). Online Learning and Online Convex Optimization. Foundations and Trends in Machine Learning, 4(2), 107–194.	Bromley, J., Guyon, I., LeCun, Y., Säckinger, E., & Shah, R. (1993). Signature verification using a 'Siamese' time delay neural network. Advances in Neural Information Processing Systems, 6.
Emërtime të tjera	incremental learning, sequential learning, streaming learning, online machine learning	twin network, Siamese neural network, contrastive metric network, Siyam ağı
Të lidhura	6	1
Përmbledhja	Online learning is a machine learning paradigm in which a model is updated incrementally as each new data point arrives, rather than being trained once on a fixed dataset. It is essential when data streams continuously, storage is limited, or the underlying distribution shifts over time. Theoretical performance is measured by cumulative regret relative to the best fixed predictor in hindsight.	A Siamese network is a deep architecture with two (or more) identical, weight-sharing branches that map inputs into an embedding space where similar inputs land close together and dissimilar ones far apart. Introduced by Bromley, LeCun, and colleagues in 1993 for signature verification and revived by Koch et al. (2015) for one-shot image recognition, it learns a similarity metric rather than fixed class labels, making it ideal for verification, matching, and few-shot tasks.