Accession Number : ADA137476

Title :   Distributed Optimization Algorithms with Communications,

Corporate Author : MASSACHUSETTS INST OF TECH CAMBRIDGE LAB FOR INFORMATION AND DECISION SYSTEMS

Personal Author(s) : Tsitsiklis,J ; Athans,M

PDF Url : ADA137476

Report Date : Nov 1983

Pagination or Media Count : 4

Abstract : This document discusses the convergence properties of asynchronous distributed iterative optimization algorithms, tolerating communication delays. The authors focus on a gradient-type algorithm for minimizing an additive cost function and present sufficient conditions for convergence. They view such an algorithm as a model of adjustment of the decisions of decision makers in an organization and we suggest that our results can be interpreted as guidelines for designing the information flows in an organization. (Author)

Descriptors :   *Algorithms, *Mathematical models, *Communications traffic, *Delay, Decision making, Asynchronous systems, Convergence, Distribution, Optimization, Iterations, Stochastic processes

Subject Categories : Statistics and Probability
      Non-radio Communications

Distribution Statement : APPROVED FOR PUBLIC RELEASE