For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Partial Label Metric Learning Based on Statistical Inference
Tian XIE Hongchang CHEN Tuosiyu MING Jianpeng ZHANG Chao GAO Shaomei LI Yuehang DING
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2020/06/01
Online ISSN: 1745-1361
Type of Manuscript: PAPER
Category: Artificial Intelligence, Data Mining
partial label learning, metric learning, statistical inference, likelihood-ratio test,
Full Text: PDF(1.5MB)>>
In partial label data, the ground-truth label of a training example is concealed in a set of candidate labels associated with the instance. As the ground-truth label is inaccessible, it is difficult to train the classifier via the label information. Consequently, manifold structure information is adopted, which is under the assumption that neighbor/similar instances in the feature space have similar labels in the label space. However, the real-world data may not fully satisfy this assumption. In this paper, a partial label metric learning method based on likelihood-ratio test is proposed to make partial label data satisfy the manifold assumption. Moreover, the proposed method needs no objective function and treats the data pairs asymmetrically. The experimental results on several real-world PLL datasets indicate that the proposed method outperforms the existing partial label metric learning methods in terms of classification accuracy and disambiguation accuracy while costs less time.