Marginal likelihoods for distributed estimation of graphical model parameters

Zhaoshi Meng, Dennis Wei, Alfred O. Hero, Ami Wiesel

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

This paper considers the estimation of graphical model parameters with distributed data collection and computation. We first discuss the use and limitations of well-known distributed methods for marginal inference in the context of parameter estimation. We then describe an alternative framework for distributed parameter estimation based on maximizing marginal likelihoods. Each node independently estimates local parameters through solving a low-dimensional convex optimization with data collected from its local neighborhood. The local estimates are then combined into a global estimate without iterative message-passing. We provide an asymptotic analysis of the proposed estimator, deriving in particular its rate of convergence. Numerical experiments validate the rate of convergence and demonstrate performance equivalent to the centralized maximum likelihood estimator.

Original languageEnglish
Title of host publication2013 5th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2013
Pages73-76
Number of pages4
DOIs
StatePublished - 2013
Event2013 5th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2013 - Saint Martin, France
Duration: 15 Dec 201318 Dec 2013

Publication series

Name2013 5th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2013

Conference

Conference2013 5th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2013
Country/TerritoryFrance
CitySaint Martin
Period15/12/1318/12/13

Fingerprint

Dive into the research topics of 'Marginal likelihoods for distributed estimation of graphical model parameters'. Together they form a unique fingerprint.

Cite this