Original scientific paper

Un upper bound for Kullback-Leibler divergence with a small number of outliers

Alexander Gofman ; Faculty of Economics, Moscow Economics National Research University, Moscow
Mark Kelbert ; Department of Mathematics, Swansea University, Singleton Park, Swansea, UK

Abstract

We establish a new upper bound for the Kullback-Leibler divergence of two discrete probability distributions which are close in a sense that typically the ratio of probabilities is nearly one and the number of outliers is small.

Keywords

Kullback-Leibler divergence; relative entropy; mutual information; information inequality

Hrčak ID:

101400

URI

https://hrcak.srce.hr/101400

Publication date:

10.5.2013.

Visits: 2.553 *