计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (12): 51-52.

• 理论研究 • 上一篇    下一篇

加权融合方法的泛化误差分解

崔丽娟1,王秋菊2,李 凯3   

  1. 1.河北大学 图书馆,河北 保定 071002
    2.河北大学 新闻传播学院,河北 保定 071002
    3.河北大学 数学与计算机学院,河北 保定 071002
  • 收稿日期:2007-11-28 修回日期:2008-02-25 出版日期:2008-04-21 发布日期:2008-04-21
  • 通讯作者: 崔丽娟

Decomposition of generalization error for weighting fusion

CUI Li-juan1,WANG Qiu-ju2,LI Kai3   

  1. 1.Library,Hebei University,Baoding,Hebei 071002,China
    2.Faculty of Journalism and Communication,Hebei University,Baoding,Hebei 071002,China
    3.School of Mathematics and Computer,Hebei University,Baoding,Hebei 071002,China
  • Received:2007-11-28 Revised:2008-02-25 Online:2008-04-21 Published:2008-04-21
  • Contact: CUI Li-juan

摘要: 机器学习的性能可以通过泛化误差表达,泛化误差越小,则该学习性能越好,反之则性能越差。为了进一步研究泛化误差的特性,通常采用泛化误差分解的方法。针对加权融合方法,并应用平方误差损失函数,给出了泛化误差的一种分解,在此基础上,进一步获得了加权融合方法的最优泛化误差分解。

关键词: 融合, 泛化误差分解, 方差, 偏差, 协方差

Abstract: The performance of machine learning may be expressed as the generalization error.To further study characteristic of generalization error,the decomposition method for generalization error is usually used.Aiming at weighting fusion method and quadratic error loss function,the decomposition equation for generalization error is given.On the basis of this,the decomposition equation for the optimal fusion method is further obtained.

Key words: fusion, decomposition of generalization error, variance, bias, covariance