Semi-Supervised Learning
A paradigm between supervised and unsupervised learning that trains models using a small amount of labeled data together with a large amount of unlabeled data. It is suited to scenarios where labeling is costly.
Core Facts
Timeline (last 90 days)
In semi-supervised learning in machine learning, unlabeled data usually requires additional assumptions or prior knowledge to be utilized.
Common methods in academia and industry for addressing few-shot defect detection include anomaly detection (such as autoencoder reconstruction error methods), data augmentation and synthetic defect generation, transfer learning, and semi-supervised/self-supervised learning.
The teacher-student architecture is a classic paradigm in semi-supervised learning, in which the teacher model generates pseudo-labels for unlabeled samples and the student model is trained using those pseudo-labels as supervision signals.
All Facts (4)
Machine learning is divided into three major categories: supervised learning, unsupervised learning, and semi-supervised learning.
80%UnverifiedIn semi-supervised learning in machine learning, unlabeled data usually requires additional assumptions or prior knowledge to be utilized.
50%UnverifiedCommon methods in academia and industry for addressing few-shot defect detection include anomaly detection (such as autoencoder reconstruction error methods), data augmentation and synthetic defect generation, transfer learning, and semi-supervised/self-supervised learning.
50%UnverifiedThe teacher-student architecture is a classic paradigm in semi-supervised learning, in which the teacher model generates pseudo-labels for unlabeled samples and the student model is trained using those pseudo-labels as supervision signals.
50%