Evaluating the Generalization Performance of SVMs Based on the Advanced Joachims Bound

SONG Xiao-shan; JIANG Xiao-yu; WANG Xi; YAO Jun

您当前的位置：

首页 >

文章列表页 >

Evaluating the Generalization Performance of SVMs Based on the Advanced Joachims Bound

更新时间：2025-07-16

- Evaluating the Generalization Performance of SVMs Based on the Advanced Joachims Bound
- Acta Electronica Sinica Vol. 39, Issue 6, Pages: 1379-1383(2011)
- 作者机构：
  
  装甲兵工程学院控制工程系,北京,100072
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP18
- Published：2011
- 稿件说明：
移动端阅览
SONG Xiao-shan, JIANG Xiao-yu, WANG Xi, et al. Evaluating the Generalization Performance of SVMs Based on the Advanced Joachims Bound[J]. Acta Electronica Sinica, 2011, 39(6): 1379-1383.
DOI：

SONG Xiao-shan, JIANG Xiao-yu, WANG Xi, et al. Evaluating the Generalization Performance of SVMs Based on the Advanced Joachims Bound[J]. Acta Electronica Sinica, 2011, 39(6): 1379-1383. DOI：

摘要

留一法(Leave One Out

LOO)错误率是评价支持向量机(Support Vector Machine

SVM)性能最准确方法

LOO错误率越小

SVM泛化性能越好.但LOO实现起来较为费时.因此人们提出了LOO错误率的各种上界作为近似

最有代表性的是Joachims上界和Jaakkola-Haussler上界.基于LOO上界的SVM泛化性能评价方法不但较为准确

而且耗时大大减小.本文首先证明了在径向基函数(Radial Basis Function

RBF)为核函数的SVM中

Joachims上界和Jaakkola-Haussler上界是等价的;其次对Joachims上界进行理论分析

指出了其不足之处

并予以改进

得到了改进的Joachims上界;最后通过实验对LOO错误率、Jaakkola-Haussler上界、Joachims上界和改进的Joachims上界进行了比较

结果显示改进的Joachims上界比Jaakkola-Haussler上界和Joachims上界更加接近LOO错误率

是一种更加准确的SVM泛化性能评价方法.

Abstract

LOO (Leave One Out) is commonly used to evaluate the generalization performance of an SVM (Support Vector Machine)

the disadvantage of which is time consuming.In order to decrease the time cost

several LOO bounds are proposed.The most famous bounds are Joachims bound and Jaakkola-Haussler bound.Based on the two bounds

the generalization performance of an SVM can be evaluated properly with decreased time cost.This paper gives the proof of the equivalence of the two bounds in an SVM with RBF (Radial Basis Function) kernel

analyzes the two bounds theoretically

proposes an advanced Joachims bound

compares the LOO error

Joachims bound

Jaakkola-Haussler bound and the advanced Joachims bound by simulated experiments.Results show that the advanced Joachims bound is closer to the LOO error

and is a better method to evaluate the generalization performance of an SVM.

关键词

Keywords

references

Views

1130

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Structure-Wise Feature Reconstruction for Hyperspectral Image Classification

Patch Verification Method Integrating Sentence Embedding Model and Code Features

Fault Diagnosis of Railway Point Machine Based on Improved Time- Domain Multiscale Dispersion Entropy and Support Vector Machine

Radar Polarization Modulation Countermeasures for Combined Corner Reflector: Anti Diluted Jamming

Image Classification of Landing Landform Based on Wavelet Transform and Deep Network

Related Author

XING Chang-da

WANG Mei-ling

XU Yong-chang

WANG Zhi-sheng

JIANG Ting-ting

JIANG Shu-juan

HAN Wei

CAO Yuan

Related Institution

School of Information and Control Engineering, China University of Mining and Technology

College of Compute Science and Technology, Nanjing University of Aeronautics and Astronautics

College of Automation Engineering, Nanjing University of Aeronautics and Astronautics

School of Computer Science & Technology， China University of Mining and Technology

National Engineering Research Center of Rail Transportation Operation and Control System, Beijing Jiaotong University

⁰