← World of AI
Machine Learning
Self-Supervised Learning
Learning useful representations from labels automatically created from the data.
Self-supervised learning creates training signals directly from the structure of unlabelled data. A model may predict masked words, missing image regions, future frames or whether two transformed examples came from the same source.
The model first learns general representations through pretraining and can later be adapted to specific tasks. This approach is central to modern language models, vision models and multimodal foundation models.
JOIN NOW
Begin the first module
It is free, it is the real curriculum, and if it is not for you, you have lost nothing but an evening.
Join any time · Build AI skills at your pace