Robust Graph-Based Semisupervised Learning for Noisy Labeled Data via Maximum Correntropy Criterion

Semisupervised learning (SSL) methods have been proved to be effective at solving the labeled samples shortage problem by using a large number of unlabeled samples together with a small number of labeled samples. However, many traditional SSL methods may not be robust with too much labeling noisy data. To address this issue, in this paper, we propose a robust graph-based SSL method based on maximum correntropy criterion to learn a robust and strong generalization model. In detail, the graph-based SSL framework is improved by imposing supervised information on the regularizer, which can strengthen the constraint on labels, thus ensuring that the predicted labels of each cluster are close to the true labels. Furthermore, the maximum correntropy criterion is introduced into the graph-based SSL framework to suppress labeling noise. Extensive image classification experiments prove the generalization and robustness of the proposed SSL method.

I. Introduction

Collecting labeled examples can be expensive and difficult in many data analysis and mining applications, such as medical diagnosis and analysis, human action recognition, image retrieval, and remote sensing image classification [1]–[5]. However, there is an abundance of available unlabeled data. To alleviate this issue, researchers have tried to use plentiful unlabeled data together with limited labeled data, and thus developed semisupervised learning (SSL) methods [6]. It has been confirmed that SSL not only avoids wasting data sources but also improves upon the generalization capability of traditional supervised learning methods.