基于模糊随机样本的学习过程一致收敛速度的界  

Bounds on the Rate of Uniform Convergence of Learning Process Based on Fuzzy Random Samples

作　　者：李俊华[1] 

机构地区：[1]河北大学数学与信息科学学院,河北保定071002

出　　处：《模糊系统与数学》2016年第6期172-175,共4页Fuzzy Systems and Mathematics

基　　金：国家自然科学基金资助项目(11201110);河北省教育厅科研项目(QN20131055)

摘　　要：学习过程一致收敛速度的界决定了学习机器的推广能力,在统计学习理论中起着很重要的作用。以刘宝碇提出的模糊随机变量的概念和基于模糊随机样本的学习理论关键定理为基础,讨论了基于模糊随机样本的学习过程一致收敛速度的界,并给出了这些界与函数容量之间的关系。Bounds on the rate of uniform convergence of learning process decided the generalization ability of learning machine and played important role in statistical learning theory. Based on the definition of fuzzy random variable provided by Liu and the key theorem of learning theory based on fuzzy random samples, the bounds on the rate of uniform convergence of learning process are discussed, and the relation between these bounds and the capacity of the set of functions is provided.

关 键 词：模糊随机变量 模糊随机经验风险最小化原则 一致收敛速度的界 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]