AIRiskAware
AI Governance Glossary
Governance Practice

What Is Semi-Supervised Learning?

Semi-Supervised Learning is a machine-learning approach that trains on a small amount of labelled data together with a larger amount of unlabelled data.

Definition

Semi-Supervised Learninga machine-learning approach that trains on a small amount of labelled data together with a larger amount of unlabelled data.

Semi-supervised learning is common where labelling is expensive but raw data is plentiful. It can improve performance over using labelled data alone, but it inherits the data-governance concerns of both approaches: the quality of the few labels and the representativeness of the much larger unlabelled set.

Source: ISO/IEC 22989:2022 (AI concepts and terminology)

Plain-language explanation

Semi-supervised learning is common where labelling is expensive but raw data is plentiful. It can improve performance over using labelled data alone, but it inherits the data-governance concerns of both approaches: the quality of the few labels and the representativeness of the much larger unlabelled set.

Primary source: ISO/IEC 22989:2022 (AI concepts and terminology)

Related terms

Supervised Learning Unsupervised Learning Machine Learning

See where you stand on AI governance

Take the free 7-question maturity assessment and get a personalised action plan.

Free assessment — 3 minutes →