Unsupervised learning
비지도학습
Finding structure or groupings in data without any labels.
Clustering similar items and reducing data to a few axes are the classic examples.
No labels makes data easy to gather, but interpretation stays human. A cluster existing does not make it meaningful.
Pretraining large models also runs without human labels, predicting the next word in raw text. That variant is often called self supervised learning.
- UsesSegmenting customers, spotting outliers, summarizing data.
- CautionHumans name the clusters.