Semi-Supervised Learning
A training approach that combines a small amount of labeled data with a large amount of unlabeled data.
Overview
Semi-supervised learning is a machine learning paradigm that sits between supervised learning (all labeled data) and unsupervised learning (no labeled data). It leverages a small set of labeled examples alongside a much larger pool of unlabeled data to improve model performance.
Key Details
This approach is particularly valuable in real-world scenarios where labeling data is expensive or time-consuming. Common techniques include pseudo-labeling, consistency regularization, and self-training, where the model generates labels for unlabeled data and retrains on its own predictions.
Related Concepts
supervised learning • unsupervised learning • self supervised learning