On Rényi measures and hypothesis testing

On Rényi measures and hypothesis testing
复制标题

关于 Rényi 措施和假设检验

DOI:
--
复制
发表时间:
2011
期刊:
IEEE International Symposium on Information Theory. Proceedings
影响因子:
--
通讯作者:
O. Shayevitz
O. Shayevitz
中科院分区:
--
文献类型:
--
作者:
O. Shayevitz

文献摘要

被引文献

相似文献

我们通过它们的Shannon对应项给出了各种Rényi信息度量的变分刻画,并证明了如何通过后者的相关性质从第一原理恢复前者的性质。在此基础上,我们对两传感器复合假设检验框架中的Rényi发散给出了一种新的操作解释。
We provide a variational characterization for the various Rényi information measures via their Shannon counterparts, and demonstrate how properties of the former can be recovered from first principle via the associated properties of the latter. Motivated by this characterization, we give a new operational interpretation for the Rényi divergence in a two-sensor composite hypothesis testing framework.