Kernel-based learning machines research overview
Kernel-based learning machines research overview
复制标题
DOI：
--
复制
发表时间：
2010
期刊：
影响因子：
--
通讯作者：
Wang Sheng-cai
中科院分区：
文献类型：
--
作者：
Wang Sheng-cai
认为核方法的研究正在成为近年来机器学习领域的一个新的热点。核方法已成功地应用于许多领域。但作为一种新技术，基于核的学习机仍然存在许多局限性。本文首先分析了基于核学习机的基本思想，然后介绍了近年来由监督学习和无监督学习发展起来的几种核学习机。最后指出了核学习机的关键问题和未来的研究方向。期望全面理解核方法。
It is believed that the study of kernel methods is becoming a new hot area in the field of machine learning recently. Kernel methods have been applied in many fields successfully. But as a new technique,there are still plenty of limitations in kernel-based learning machines. This paper firstly analyzed the basic ideas of kernel-based learning machines then provided an introduction to some kinds of kernel learning machines developed recently from supervised learning and unsupervised learning. Pointed out the key issues and the future research directions on kernel learning machines finally. It is expected to understand kernel methods comprehensively.