Accession Number : AD1022782

Title :   Distributed Information Fusion through Advanced Multi-Agent Control

Descriptive Note : Technical Report,14 May 2014,13 May 2016


Personal Author(s) : Bishop, Adrian

Full Text :

Report Date : 09 Sep 2016

Pagination or Media Count : 5

Abstract : Distributed consensus in the Wasserstein metric space of probability measures was the primary topic of investigation under this project. Convergence of each agent's (or nodes) measure to a common probability measure is proven under a weak network connectivity condition. The common measure reached at each agent is one minimizing a weighted sum of its Wasserstein distance to all initial agent measures. This measure is known as the Wasserstein barycenter. Special cases involving Gaussian measures, empirical measures, and time-invariant network topologies are considered, where convergence rates and average-consensus results are given. This algorithm has potential applicability in computer vision, machine learning and distributed estimation, etc. A number of other topics in distributed and Monte-Carlo estimation were also considered including: distributed information fusion under unknown correlations; large-scale sequential Monte-Carlo methods; optimal controller approximation via Monte-Carlo methods; score and information matrix approximation via sequential Monte-Carlo methods.

Descriptors :   NETWORK TOPOLOGY , algorithms , GAUSSIAN PROCESSES , probability , Multiagent systems , Data fusion

Subject Categories : Numerical Mathematics
      Statistics and Probability
      Computer Programming and Software
      Computer Systems

Distribution Statement : APPROVED FOR PUBLIC RELEASE