Oral
Classification from Positive, Unlabeled and Biased Negative Data
Yu-Guan Hsieh · Gang Niu · Masashi Sugiyama

Thu Jun 13th 09:20 -- 09:25 AM @ Room 103

In binary classification, there are situations where negative (N) data are too diverse to be fully labelled and that is when positive-unlabeled (PU) learning comes into play. However, collecting a non-representative N set that contains only a small portion of all possible N data can be much easier in many practical situations. This paper studies a novel classification framework which incorporates such biased N (bN) data in PU learning. We provide a method based on empirical risk minimization to address this PUbN classification problem. Our approach can be regarded as a novel example-reweighting algorithm, with the weight of each example computed through a preliminary step that draws inspiration from PU learning. We also derive an estimation error bound for the proposed method. Experimental results demonstrate the effectiveness of our algorithm in not only PUbN learning scenarios but also ordinary PU leaning scenarios on several benchmark datasets.

Author Information

Yu-Guan Hsieh (École normale supérieure)
Gang Niu (RIKEN)
Masashi Sugiyama (RIKEN / The University of Tokyo)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors