Distributed optimization over time-varying directed graphs

Angelia Nedich, Alex Olshevsky

Research output: Contribution to journalArticle

236 Citations (Scopus)

Abstract

We consider distributed optimization by a collection of nodes, each having access to its own convex function, whose collective goal is to minimize the sum of the functions. The communications between nodes are described by a time-varying sequence of directed graphs, which is uniformly strongly connected. For such communications, assuming that every node knows its out-degree, we develop a broadcast-based algorithm, termed the subgradient-push, which steers every node to an optimal value under a standard assumption of subgradient boundedness. The subgradient-push requires no knowledge of either the number of agents or the graph sequence to implement. Our analysis shows that the subgradient-push algorithm converges at a rate of O(ln t/√t). The proportionality constant in the convergence rate depends on the initial values at the nodes, the subgradient norms and, more interestingly, on both the speed of the network information diffusion and the imbalances of influence among the nodes.

Original languageEnglish (US)
Article number6930814
Pages (from-to)601-615
Number of pages15
JournalIEEE Transactions on Automatic Control
Volume60
Issue number3
DOIs
StatePublished - Mar 1 2015
Externally publishedYes

Fingerprint

Directed graphs
Communication

Keywords

  • Time-varying
  • UAVs

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Computer Science Applications

Cite this

Distributed optimization over time-varying directed graphs. / Nedich, Angelia; Olshevsky, Alex.

In: IEEE Transactions on Automatic Control, Vol. 60, No. 3, 6930814, 01.03.2015, p. 601-615.

Research output: Contribution to journalArticle

@article{e738485ecdf244eb9b32667ddd6f1a48,
title = "Distributed optimization over time-varying directed graphs",
abstract = "We consider distributed optimization by a collection of nodes, each having access to its own convex function, whose collective goal is to minimize the sum of the functions. The communications between nodes are described by a time-varying sequence of directed graphs, which is uniformly strongly connected. For such communications, assuming that every node knows its out-degree, we develop a broadcast-based algorithm, termed the subgradient-push, which steers every node to an optimal value under a standard assumption of subgradient boundedness. The subgradient-push requires no knowledge of either the number of agents or the graph sequence to implement. Our analysis shows that the subgradient-push algorithm converges at a rate of O(ln t/√t). The proportionality constant in the convergence rate depends on the initial values at the nodes, the subgradient norms and, more interestingly, on both the speed of the network information diffusion and the imbalances of influence among the nodes.",
keywords = "Time-varying, UAVs",
author = "Angelia Nedich and Alex Olshevsky",
year = "2015",
month = "3",
day = "1",
doi = "10.1109/TAC.2014.2364096",
language = "English (US)",
volume = "60",
pages = "601--615",
journal = "IEEE Transactions on Automatic Control",
issn = "0018-9286",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

TY - JOUR

T1 - Distributed optimization over time-varying directed graphs

AU - Nedich, Angelia

AU - Olshevsky, Alex

PY - 2015/3/1

Y1 - 2015/3/1

N2 - We consider distributed optimization by a collection of nodes, each having access to its own convex function, whose collective goal is to minimize the sum of the functions. The communications between nodes are described by a time-varying sequence of directed graphs, which is uniformly strongly connected. For such communications, assuming that every node knows its out-degree, we develop a broadcast-based algorithm, termed the subgradient-push, which steers every node to an optimal value under a standard assumption of subgradient boundedness. The subgradient-push requires no knowledge of either the number of agents or the graph sequence to implement. Our analysis shows that the subgradient-push algorithm converges at a rate of O(ln t/√t). The proportionality constant in the convergence rate depends on the initial values at the nodes, the subgradient norms and, more interestingly, on both the speed of the network information diffusion and the imbalances of influence among the nodes.

AB - We consider distributed optimization by a collection of nodes, each having access to its own convex function, whose collective goal is to minimize the sum of the functions. The communications between nodes are described by a time-varying sequence of directed graphs, which is uniformly strongly connected. For such communications, assuming that every node knows its out-degree, we develop a broadcast-based algorithm, termed the subgradient-push, which steers every node to an optimal value under a standard assumption of subgradient boundedness. The subgradient-push requires no knowledge of either the number of agents or the graph sequence to implement. Our analysis shows that the subgradient-push algorithm converges at a rate of O(ln t/√t). The proportionality constant in the convergence rate depends on the initial values at the nodes, the subgradient norms and, more interestingly, on both the speed of the network information diffusion and the imbalances of influence among the nodes.

KW - Time-varying

KW - UAVs

UR - http://www.scopus.com/inward/record.url?scp=84923608183&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923608183&partnerID=8YFLogxK

U2 - 10.1109/TAC.2014.2364096

DO - 10.1109/TAC.2014.2364096

M3 - Article

VL - 60

SP - 601

EP - 615

JO - IEEE Transactions on Automatic Control

JF - IEEE Transactions on Automatic Control

SN - 0018-9286

IS - 3

M1 - 6930814

ER -