Rademacher complexity in Neyman-Pearson classification

详细信息    查看全文

• 作者：Min Han ; Di Rong Chen and Zhao Xu Sun
• 关键词：Neyman ; Pearson lemma ; VC classes ; Rademacher complexity ; Neyman ; Pearson classification
• 刊名：Acta Mathematica Sinica
• 出版年：2009
• 出版时间：May, 2009
• 年：2009
• 卷：25
• 期：5
• 页码：855-868
• 全文大小：251.7 KB

文摘

Neyman-Pearson(NP) criterion is one of the most important ways in hypothesis testing. It is also a criterion for classification. This paper addresses the problem of bounding the estimation error of NP classification, in terms of Rademacher averages. We investigate the behavior of the global and local Rademacher averages, and present new NP classification error bounds which are based on the localized averages, and indicate how the estimation error can be estimated without a priori knowledge of the class at hand.