Large Margin Semi-supervised Learning

In classification, semi-supervised learning occurs when a large amount of unlabeled data is available with only a small number of labeled data. In such a situation, how to enhance predictability of classification through unlabeled data is the focus. In this article, we introduce a novel large margin semi-supervised learning methodology, using grouping information from unlabeled data, together with the concept of margins, in a form of regularization controlling the interplay between labeled and unlabeled data. Based on this methodology, we develop two specific machines involving support vector machines and ψ-learning, denoted as SSVM and SPSI, through difference convex programming. In addition, we estimate the generalization error using both labeled and unlabeled data, for tuning regularizers. Finally, our theoretical and numerical analyses indicate that the proposed methodology achieves the desired objective of delivering high performance in generalization, particularly against some strong performers.