Generalizing Jensen and Bregman divergences with comparative convexity
and the statistical Bhattacharyya distances with comparable means
F. Nielsen, und R. Nock. (2017)cite arxiv:1702.04877Comment: 24 pages.
Zusammenfassung
Comparative convexity is a generalization of convexity relying on abstract
notions of means. We define the Jensen divergence and the Jensen diversity from
the viewpoint of comparative convexity, and show how to obtain the generalized
Bregman divergences as limit cases of skewed Jensen divergences. In particular,
we report explicit formula of these generalized Bregman divergences when
considering quasi-arithmetic means. Finally, we introduce a generalization of
the Bhattacharyya statistical distances based on comparative means using
relative convexity.
Beschreibung
[1702.04877] Generalizing Jensen and Bregman divergences with comparative convexity and the statistical Bhattacharyya distances with comparable means
%0 Journal Article
%1 nielsen2017generalizing
%A Nielsen, Frank
%A Nock, Richard
%D 2017
%K divergences entropy information theory
%T Generalizing Jensen and Bregman divergences with comparative convexity
and the statistical Bhattacharyya distances with comparable means
%U http://arxiv.org/abs/1702.04877
%X Comparative convexity is a generalization of convexity relying on abstract
notions of means. We define the Jensen divergence and the Jensen diversity from
the viewpoint of comparative convexity, and show how to obtain the generalized
Bregman divergences as limit cases of skewed Jensen divergences. In particular,
we report explicit formula of these generalized Bregman divergences when
considering quasi-arithmetic means. Finally, we introduce a generalization of
the Bhattacharyya statistical distances based on comparative means using
relative convexity.
@article{nielsen2017generalizing,
abstract = {Comparative convexity is a generalization of convexity relying on abstract
notions of means. We define the Jensen divergence and the Jensen diversity from
the viewpoint of comparative convexity, and show how to obtain the generalized
Bregman divergences as limit cases of skewed Jensen divergences. In particular,
we report explicit formula of these generalized Bregman divergences when
considering quasi-arithmetic means. Finally, we introduce a generalization of
the Bhattacharyya statistical distances based on comparative means using
relative convexity.},
added-at = {2019-12-11T14:33:13.000+0100},
author = {Nielsen, Frank and Nock, Richard},
biburl = {https://www.bibsonomy.org/bibtex/29957acff015382dba208f035f3d3c4e9/kirk86},
description = {[1702.04877] Generalizing Jensen and Bregman divergences with comparative convexity and the statistical Bhattacharyya distances with comparable means},
interhash = {b76bd6cf3b6ff0647232a7e552461b09},
intrahash = {9957acff015382dba208f035f3d3c4e9},
keywords = {divergences entropy information theory},
note = {cite arxiv:1702.04877Comment: 24 pages},
timestamp = {2019-12-11T14:33:13.000+0100},
title = {Generalizing Jensen and Bregman divergences with comparative convexity
and the statistical Bhattacharyya distances with comparable means},
url = {http://arxiv.org/abs/1702.04877},
year = 2017
}