A robust ensemble approach to learn from positive and unlabeled data using SVM base models

We present a novel approach to learn binary classifiers when only positive and unlabeled instances are available (PU learning). This problem is routinely cast as a supervised task with label noise in the negative set. We use an ensemble of SVM models trained on bootstrap subsamples of the training data for increased robustness against label noise. The approach can be considered in a bagging framework which provides an intuitive explanation for its mechanics in a semi-supervised setting. We compared our method to stateof-the-art approaches in a simulation based on the MNIST digit recognition data set. The included benchmark comprises three settings with increasing label noise: (i) fully supervised, (ii) PU learning and (iii) PU learning with false positives. Our approach shows a marginal improvement over existing methods in the second setting and a significant improvement in the third.