An information-theoretic proof of a finite de finetti theorem


Type
Article
Change log
Authors
Gavalakis, L 
Kontoyiannis, Ioannis  ORCID logo  https://orcid.org/0000-0001-7242-6375
Abstract

A finite form of de Finetti's representation theorem is established using elementary information-theoretic tools: The distribution of the first k random variables in an exchangeable binary vector of length nk is close to a mixture of product distributions. Closeness is measured in terms of the relative entropy and an explicit bound is provided.

Description
Keywords
cs.IT, cs.IT, math.IT, math.PR
Journal Title
Electronic Communications in Probability
Conference Name
Journal ISSN
1083-589X
1083-589X
Volume Title
26
Publisher
Institute of Mathematical Statistics