A Class of Lower Bounds for Bayesian Risk with a Bregman Loss
A Class of Lower Bounds for Bayesian Risk with a Bregman Loss
复制标题
一类具有 Bregman 损失的贝叶斯风险下界
DOI：
--
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
L. F. I. H. Vincent Poor
中科院分区：
文献类型：
--
作者：
Member Ieee Alex Dytso;Member Ieee Michael Fauß;L. F. I. H. Vincent Poor
当潜在损失函数是Bregman发散时，证明了一般类的贝叶斯下界。这类可以被认为是一个扩展的Weinstein-Weiss家庭的均方误差的界限，并依赖于找到一个变分表征贝叶斯风险。该方法允许导出特定于给定Bregman发散的Cramér-Rao界的版本。在泊松噪声环境下，新的界限的有效性进行了评估。
A general class of Bayesian lower bounds when the underlying loss function is a Bregman divergence is demonstrated. This class can be considered as an extension of the Weinstein–Weiss family of bounds for the mean squared error and relies on finding a variational characterization of Bayesian risk. The approach allows for the derivation of a version of the Cramér–Rao bound that is specific to a given Bregman divergence. The effectiveness of the new bound is evaluated in the Poisson noise setting.
泊松噪声的估计：条件均值估计器的属性
影响因子：
2.5
作者：
Dytso, Alex;Vincent Poor, H.
通讯作者：
Vincent Poor, H.