Large-Scale Semi-Supervised Learning
Weston Jason · NATO science for peace and security series. D, Information and communication security · 2008
Labeling data is expensive, whilst unlabeled data is often abundant and cheap to collect. Semi-supervised learning algorithms that use both types of data can perform significantly better than supervised algorithms that use labeled data alone. However, for such gains to be observed, the amount of unlabeled data trained on should be relatively large. Therefore, making semi-supervised algorithms scalable is paramount. In this work we review several recent techniques for semi-supervised learning, and methods for improving the scalability of these algorithms.