计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (10): 40-42.

• 理论研究 • 上一篇    下一篇

一种基于粗糙变量的学习算法的基础研究

董开坤1,刘 杨1,刘 扬1,胡仕成1,刘慧霞2   

  1. 1.哈尔滨工业大学(威海) 计算机科学与技术学院,山东 威海 264209
    2.青岛大学 数学科学学院,山东 青岛 266071
  • 收稿日期:2007-11-20 修回日期:2008-01-28 出版日期:2008-04-01 发布日期:2008-04-01
  • 通讯作者: 董开坤

Fundamental reaching of kind of learning algorithm based on rough variable

DONG Kai-kun1,LIU Yang1,LIU Yang1,HU Shi-cheng1,LIU Hui-xia2   

  1. 1.Department of Computer Science & Technology,Harbin Institute of Technology at Weihai,Weihai,Shandong 264209,China
    2.Mathematics Science College,Qingdao University,Qingdao,Shandong 266071,China
  • Received:2007-11-20 Revised:2008-01-28 Online:2008-04-01 Published:2008-04-01
  • Contact: DONG Kai-kun

摘要: 支持向量机目前已成为机器学习领域新的研究热点,而统计学习理论中的关键定理为支持向量机等的研究提供了重要的理论基础。提出了粗糙经验风险最小化原则,提出并证明了一种基于粗糙变量的学习理论的关键定理,为研究粗糙支持向量机等应用性研究提供了依据。

关键词: 向量机, 信赖理论, 信赖统计, 粗糙经验风险最小化原则, 关键定理

Abstract: Support vector machine is one of the new hotspots in the area of machine learning now.The key theorem of statistical learning theory provides a theoretical basis for its research.The rough empirical risk minimization principle is proposed,and the key theorem of learning theory based on rough variable is advanced in the paper.This provides evidence for the application of rough support vector machine etc.

Key words: vector machines, trust theory, trust statistics, rough empirical risk minimization principle, the key theorem