カンファレンス (国際) Covariate Shift Adaptation on Learning from Positive and Unlabeled Data
Tomoya Sakai (NEC Corp.), Nobuyuki Shimizu
The Thirty-Third AAAI Conference on Artificial Intelligence (AAAI-19)
The goal of binary classification is to identify whether an in- put sample belongs to positive or negative classes. Usually, supervised learning is applied to obtain a classification rule, but in real-world applications, it is conceivable that only pos- itive and unlabeled data are accessible for learning, which is called learning from positive and unlabeled data (PU learn- ing). Furthermore, in practice, data distributions are likely to differ between training and testing due to, for example, time variation and domain shift. The covariate shift is a dataset shift situation, where distributions of covariates (inputs) dif- fer between training and testing, but the input-output rela- tion is the same. In this paper, we address the PU learning problem under the covariate shift. We propose an importance- weighted PU learning method and reveal in which situations the importance-weighting is necessary. Moreover, we derive the convergence rate of the proposed method under mild con- ditions and experimentally demonstrate its effectiveness.