期刊
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
卷 33, 期 4, 页码 1632-1643出版社
IEEE COMPUTER SOC
DOI: 10.1109/TKDE.2019.2947040
关键词
Training; Correlation; Buildings; Supervised learning; Pose estimation; Transforms; Label enhancement; label distribution learning; multi-label learning; learning with ambiguity
类别
资金
- National Key Research & Development Plan of China [2017YFB1002801]
- National Science Foundation of China [61622203]
- Jiangsu Natural Science Funds for Distinguished Young Scholar [BK20140022]
- Collaborative Innovation Center of Novel Software Technology and Industrialization
- Collaborative Innovation Center of Wireless Communications Technology
Label distribution learning covers a certain number of labels and the process of recovering label distributions can enhance the supervision information in training sets, leading to better learning performance.
Label distribution is more general than both single-label annotation and multi-label annotation. It covers a certain number of labels, representing the degree to which each label describes the instance. The learning process on the instances labeled by label distributions is called label distribution learning (LDL). Unfortunately, many training sets only contain simple logical labels rather than label distributions due to the difficulty of obtaining the label distributions directly. To solve this problem, one way is to recover the label distributions from the logical labels in the training set via leveraging the topological information of the feature space and the correlation among the labels. Such process of recovering label distributions from logical labels is defined as label enhancement (LE), which reinforces the supervision information in the training sets. This paper proposes a novel LE algorithm called Graph Laplacian Label Enhancement (GLLE). Experimental results on one artificial dataset and fourteen real-world LDL datasets show clear advantages of GLLE over several existing LE algorithms. Furthermore, experimental results on eleven multi-label learning datasets validate the advantage of GLLE over the state-of-the-art multi-label learning approaches.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据