A Marginal Characterization of Entropy Functions for Conditional Mutually Independent Random Variables (with Application to Wyner's Common Information) Conference Paper

abstract

• 2015 IEEE. We prove that by imposing a conditional mutual independence constraint and a marginalisation constraint, the almost entropic region can be completely characterised by Shannon-type information inequalities. Such a property is applied to obtain an explicit lower bound on the generalised Wyner common information.

name of conference

• 2015 IEEE International Symposium on Information Theory (ISIT)

authors

• Liu, Tie

published proceedings

• 2015 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT)

author list (cited authors)

• Chen, Q. i., Cheng, F., Liu, T., & Yeung, R. W.

citation count

• 3

complete list of authors

• Chen, Qi||Cheng, Fan||Liu, Tie||Yeung, Raymond W

publication date

• January 2015

publisher

• Institute of Electrical and Electronics Engineers (IEEE)  Publisher

published in

• IEEE International Symposium on Information Theory - Proceedings  Journal

Digital Object Identifier (DOI)

• 10.1109/isit.2015.7282600

International Standard Book Number (ISBN) 13

• 9781467377041

start page

• 974

end page

• 978

volume

• 2015-June