Classifier Ensemble with Unlabeled Data

Ensemble learning aims to improve generalization ability by using multiple base learners. It is well-known that to construct a good ensemble, the base learners should be accurate as well as diverse. In this paper, unlabeled data is exploited to facilitate ensemble learning by helping augment the diversity among the base learners. Specifically, a semi-supervised ensemble method named Sealed is proposed. Unlike existing semi-supervised ensemble methods where error-prone pseudo-labels are estimated for unlabeled data, Sealed works by maximizing accuracies of base learners on labeled data and maximizing diversity among them on labeled as well as unlabeled data. Experiments show that Sealed can effectively utilize unlabeled data for ensemble learning and is highly competitive to wellestablished semi-supervised ensemble methods.