| A Measure of Learning Model Complexity by VC Dimension |
| |
| 作者姓名： | WANGWen－jian ZHANGLi－xia 等 |
| |
| 作者单位： | WANG Wen-jian1，ZHANG Li-xia2，XU Zong-ben11.Institute for Information and System Science，Faculty of Science，Xi′an Jiaotong University，Xi′an 710049，China2.Department of Computer Science，Henan Normal University，Xinxiang 453002，China |
| |
| 摘 要： | 1 IntroductionMeasure of model complexity is important for understanding how difficult a modelingproblem is and for optimizing model performance.Modeling,in all its varied forms,is theprocess of logical induction. A key question regarding induction can be stated as:whatmust one know a priori about an unknown functional dependency in order to estimate itonthe basis of observations?1 ] . In the fields of engineering and computer science,thisquestion is known as the learning problem. Recently,…
|
| 关 键 词： | 学习模型 复杂性 统计学习理论 建模 度量 系统论 |
A Measure of Learning Model Complexity by VC Dimension |
| |
| WANG Wen-jian,ZHANG Li-xia,XU Zong-ben.A Measure of Learning Model Complexity by VC Dimension[J].Journal of Systems Science and Systems Engineering,2002,11(4):455-461. |
| |
| Authors: | WANG Wen-jian ZHANG Li-xia XU Zong-ben |
| |
| Abstract: | When developing models there is always a trade-off between model complexity and model fit. In this paper, a measure of learning model complexity based on VC dimension is presented, and some relevant mathematical theory surrounding the derivation and use of this metric is summarized. The measure allows modelers to control the amount of error that is returned from a modeling system and to state upper bounds on the amount of error that the modeling system will return on all future, as yet unseen and uncollected data sets. It is possible for modelers to use the VC theory to determine which type of model more accurately represents a system. |
| |
| Keywords: | VC dimension learning model complexity statistical learning theory modeling |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
