arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1502.03505cs.LG
收藏

对称正定矩阵的有监督LogEuclidean度量学习

Supervised LogEuclidean Metric Learning for Symmetric Positive Definite Matrices

The University of Tokyo(东京大学)

Florian Yger, Masashi Sugiyama

摘要:

度量学习已被证明是非常有效的,以提高最近邻分类的性能。在本文中,我们解决了对称正定(SPD)矩阵,如协方差矩阵,在许多现实世界的应用中出现的度量学习的问题。在欧几里得几何下对SPD矩阵简单地使用标准的马氏度量学习方法是不合适的,因为SPD矩阵的差可以是非SPD矩阵,因此所获得的解可能是不可解释的。为了解决这个问题,我们建议使用适当参数化的对数欧几里德距离,并优化相对于内核-目标对齐的度量,这是内核学习的监督标准。然后利用黎曼几何求解所得到的非平凡优化问题。最后,我们通过实验证明了我们的LogEuclidean度量学习算法对EEG信号和纹理补丁的真实分类任务的有用性。

英文摘要:

Metric learning has been shown to be highly effective to improve the performance of nearest neighbor classification. In this paper, we address the problem of metric learning for Symmetric Positive Definite (SPD) matrices such as covariance matrices, which arise in many real-world applications. Naively using standard Mahalanobis metric learning methods under the Euclidean geometry for SPD matrices is not appropriate, because the difference of SPD matrices can be a non-SPD matrix and thus the obtained solution can be uninterpretable. To cope with this problem, we propose to use a properly parameterized LogEuclidean distance and optimize the metric with respect to kernel-target alignment, which is a supervised criterion for kernel learning. Then the resulting non-trivial optimization problem is solved by utilizing the Riemannian geometry. Finally, we experimentally demonstrate the usefulness of our LogEuclidean metric learning algorithm on real-world classification tasks for EEG signals and texture patches.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑