计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (20): 45-47.DOI: 10.3778/j.issn.1002-8331.2008.20.013

• 理论研究 • 上一篇    下一篇

模糊样本下学习问题的研究

刘 杨,刘 扬,胡仕成   

  1. 哈尔滨工业大学(威海) 计算机科学与技术学院 国家计算机信息内容安全重点实验室,山东 威海 264209
  • 收稿日期:2008-03-19 修回日期:2008-05-19 出版日期:2008-07-11 发布日期:2008-07-11
  • 通讯作者: 刘 杨

Research of learning problem with fuzzy samples

LIU Yang,LIU Yang,HU Shi-cheng   

  1. School of Computer Science & Technology,Harbin Institute of Technology at Weihai,Weihai,Shandong 264209,China
  • Received:2008-03-19 Revised:2008-05-19 Online:2008-07-11 Published:2008-07-11
  • Contact: LIU Yang

摘要: 支持向量机是机器学习领域一个研究热点,而统计学习理论中的关键定理为支持向量机的研究提供了重要的理论基础。基于模糊样本,提出了模糊经验风险最小化原则和非平凡一致性的概念,提出并证明了基于模糊样本的学习理论的关键定理,为研究模糊支持向量机提供了依据。

Abstract: Support Vector Machine(SVM) is a focus in the area of machine learning,and the key theorem of statistical learning theory provides important basis for SVM.Fuzzy Empirical Risk Minimization(FERM) principle is proposed in the paper.The key theorem of learning theory with fuzzy samples is proposed and proven.And it provides theoretical basis for the research of fuzzy support vector machine.