方法对比

	在线自监督学习	在线学习	自监督学习
领域	机器学习	机器学习	机器学习
方法族	Machine learning	Machine learning	Machine learning
起源年份	2020s	1958–2000s	2018–2020
提出者	Multiple contributors (Gidaris, Fini et al., among others)	Rosenblatt, F.; Littlestone, N.; Shalev-Shwartz, S. (key contributors)	LeCun, Y. and community (formalized ~2018–2020)
类型	Online unsupervised representation learning	Learning paradigm (sequential model update)	Representation learning paradigm
开创性文献	Gidaris, S., Bursuc, A., Komodakis, N., Perez, P., & Cord, M. (2021). OBoW: Online Bag-of-Visual-Words Generation for Self-Supervised Learning. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 6830–6840.	Shalev-Shwartz, S. (2011). Online Learning and Online Convex Optimization. Foundations and Trends in Machine Learning, 4(2), 107–194.	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/
别名	online SSL, continual self-supervised learning, streaming self-supervised learning, incremental self-supervised learning	incremental learning, sequential learning, streaming learning, online machine learning	SSL, self-supervised pre-training, pretext-task learning, unsupervised representation learning
相关	3	6	3
摘要	Online Self-supervised Learning (online SSL) trains neural networks on unlabeled data that arrives sequentially or in streams, using automatically generated supervisory signals (pretext tasks) instead of human labels. By updating the model continuously as new data flows in, it enables perpetually evolving representations without storing the full dataset — critical for real-time systems, edge devices, and privacy-constrained settings.	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.	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.