Strongly Convex Divergences
Information Theory
2020-12-30 v1 math.IT
Probability
Statistics Theory
Statistics Theory
Abstract
We consider a sub-class of the -divergences satisfying a stronger convexity property, which we refer to as strongly convex, or -convex divergences. We derive new and old relationships, based on convexity arguments, between popular -divergences.
Cite
@article{arxiv.2009.10838,
title = {Strongly Convex Divergences},
author = {James Melbourne},
journal= {arXiv preprint arXiv:2009.10838},
year = {2020}
}
Comments
Submitted to Entropy, special issue, Divergence Measures: Mathematical Foundations and Applications in Information-Theoretic and Statistical Problems, edited by Igal Sason