节点文献
可能性空间上统计学习理论的关键定理
The Key Theorem of Statistical Learning Theory on Possibility Measure Spaces
【作者】 白云超;
【作者基本信息】 河北大学 , 应用数学, 2004, 硕士
【摘要】 本文进一步讨论了可信性测度的性质,在可能性空间上给出了车贝谢夫不等式和辛钦大数定理;并依据传统的统计学习理论在可能性空间上给出了经验风险泛函,期望风险泛函,经验风险最小化原则等新的概念,在此基础上给出并证明了统计学习理论的关键定理。
【Abstract】 In this paper, we will further discuss the property of the credibility measure and give TchebychefPs inequality and a large number theorem.On possibility measure spaces,we will give some new concepts of empirical risk functional, the expected risk functional, the empirical risk minimization inductive principle (ERM) according to theclassical statistical learning theory. At last, we will give and prove the key theorem of the statistical learning on the base of it.
【关键词】 可信性测度;
期望风险泛函;
经验风险泛函;
经验风险最小化原则;
【Key words】 credibility measure; the empirical risk functional; the expected riskfunctional; empirical risk minimize;
【Key words】 credibility measure; the empirical risk functional; the expected riskfunctional; empirical risk minimize;
- 【网络出版投稿人】 河北大学 【网络出版年期】2004年 04期
- 【分类号】O212
- 【被引频次】3
- 【下载频次】130