Foundations
Unsupervised Learning
Training a model on data with no labels, so it has to find structure and patterns on its own.
Unsupervised learning works with raw, unlabeled data and asks the model to discover the structure hidden inside it, such as natural groupings or recurring patterns. Clustering customers by behavior and compressing data into simpler representations are classic examples. Much of how large language models first learn language is self-supervised, a close relative where the data effectively labels itself by hiding and predicting its own next word.
Reference: Wikipedia: Unsupervised Learning
Related terms
Supervised Learning
Training a model on labeled examples, where each input comes paired with the correct answer.
Embedding
A way of turning words, images, or other data into lists of numbers that capture their meaning, so a computer can compare them.
Pretraining
The first, largest training phase where a model learns general patterns from a huge amount of broad data.
Machine Learning
A way of building software that learns patterns from examples instead of being told exactly what to do.