English

Maximum Entropy Functions: Approximate Gacs-Korner for Distributed Compression

Information Theory 2016-04-15 v2 math.IT

Abstract

Consider two correlated sources XX and YY generated from a joint distribution pX,Yp_{X,Y}. Their G\'acs-K\"orner Common Information, a measure of common information that exploits the combinatorial structure of the distribution pX,Yp_{X,Y}, leads to a source decomposition that exhibits the latent common parts in XX and YY. Using this source decomposition we construct an efficient distributed compression scheme, which can be efficiently used in the network setting as well. Then, we relax the combinatorial conditions on the source distribution, which results in an efficient scheme with a helper node, which can be thought of as a front-end cache. This relaxation leads to an inherent trade-off between the rate of the helper and the rate reduction at the sources, which we capture by a notion of optimal decomposition. We formulate this as an approximate G\'acs-K\"orner optimization. We then discuss properties of this optimization, and provide connections with the maximal correlation coefficient, as well as an efficient algorithm, both through the application of spectral graph theory to the induced bipartite graph of pX,Yp_{X,Y}.

Keywords

Cite

@article{arxiv.1604.03877,
  title  = {Maximum Entropy Functions: Approximate Gacs-Korner for Distributed Compression},
  author = {Salman Salamatian and Asaf Cohen and Muriel Médard},
  journal= {arXiv preprint arXiv:1604.03877},
  year   = {2016}
}

Comments

Submitted to ITW 2016

R2 v1 2026-06-22T13:31:39.845Z