English

Binary Classification from Positive-Confidence Data

Machine Learning 2018-11-29 v3 Machine Learning

Abstract

Can we learn a binary classifier from only positive data, without any negative data or unlabeled data? We show that if one can equip positive data with confidence (positive-confidence), one can successfully learn a binary classifier, which we name positive-confidence (Pconf) classification. Our work is related to one-class classification which is aimed at "describing" the positive class by clustering-related methods, but one-class classification does not have the ability to tune hyper-parameters and their aim is not on "discriminating" positive and negative classes. For the Pconf classification problem, we provide a simple empirical risk minimization framework that is model-independent and optimization-independent. We theoretically establish the consistency and an estimation error bound, and demonstrate the usefulness of the proposed method for training deep neural networks through experiments.

Keywords

Cite

@article{arxiv.1710.07138,
  title  = {Binary Classification from Positive-Confidence Data},
  author = {Takashi Ishida and Gang Niu and Masashi Sugiyama},
  journal= {arXiv preprint arXiv:1710.07138},
  year   = {2018}
}

Comments

NeurIPS 2018 camera-ready version (this paper was selected for spotlight presentation)