Automatic feature decomposition for single view co-training

One of the most successful semi-supervised learning approaches is co-training for multiview data. In co-training, one trains two classifiers, one for each view, and uses the most confident predictions of the unlabeled data for the two classifiers to "teach each other". In this paper, we extend co-training to learning scenarios without an explicit multi-view representation. Inspired by a theoretical analysis of Balcan et al. (2004), we introduce a novel algorithm that splits the feature space during learning, explicitly to encourage co-training to be successful. We demonstrate the efficacy of our proposed method in a weakly-supervised setting on the challenging Caltech-256 object recognition task, where we improve significantly over previous results by (Bergamo & Torresani, 2010) in almost all training-set size settings.