A gossip algorithm for aggregative games on graphs

Jayash Koshal, Angelia Nedich, Uday V. Shanbhag

Research output: Contribution to journalArticle

31 Citations (Scopus)

Abstract

We consider a class of games, termed as aggregative games, being played over a distributed multi-agent networked system. In an aggregative game, an agent's objective function is coupled through a function of the aggregate of all agents decisions. Every agent maintains an estimate of the aggregate and agents exchange this information over a connected network. We study the gossip-based distributed algorithm for the exchange of information and computation of equilibrium decisions of agents over the network. Our primary emphasis lies in proving the convergence of the algorithm under an assumption of a diminishing (agent-specific) step-size sequence. Under standard conditions, we establish the almost-sure convergence of the generated sequence to the unique equilibrium point.

Original languageEnglish (US)
Article number6426136
Pages (from-to)4840-4845
Number of pages6
JournalUnknown Journal
DOIs
StatePublished - 2012
Externally publishedYes

Fingerprint

Gossip
Game
Graph in graph theory
Almost Sure Convergence
Diminishing
Distributed Algorithms
Equilibrium Point
Parallel algorithms
Objective function
Estimate

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Modeling and Simulation
  • Control and Optimization

Cite this

A gossip algorithm for aggregative games on graphs. / Koshal, Jayash; Nedich, Angelia; Shanbhag, Uday V.

In: Unknown Journal, 2012, p. 4840-4845.

Research output: Contribution to journalArticle

Koshal, Jayash ; Nedich, Angelia ; Shanbhag, Uday V. / A gossip algorithm for aggregative games on graphs. In: Unknown Journal. 2012 ; pp. 4840-4845.
@article{83a97c4491d24a8e91dfb9362a5dc215,
title = "A gossip algorithm for aggregative games on graphs",
abstract = "We consider a class of games, termed as aggregative games, being played over a distributed multi-agent networked system. In an aggregative game, an agent's objective function is coupled through a function of the aggregate of all agents decisions. Every agent maintains an estimate of the aggregate and agents exchange this information over a connected network. We study the gossip-based distributed algorithm for the exchange of information and computation of equilibrium decisions of agents over the network. Our primary emphasis lies in proving the convergence of the algorithm under an assumption of a diminishing (agent-specific) step-size sequence. Under standard conditions, we establish the almost-sure convergence of the generated sequence to the unique equilibrium point.",
author = "Jayash Koshal and Angelia Nedich and Shanbhag, {Uday V.}",
year = "2012",
doi = "10.1109/CDC.2012.6426136",
language = "English (US)",
pages = "4840--4845",
journal = "Scanning Electron Microscopy",
issn = "0586-5581",
publisher = "Scanning Microscopy International",

}

TY - JOUR

T1 - A gossip algorithm for aggregative games on graphs

AU - Koshal, Jayash

AU - Nedich, Angelia

AU - Shanbhag, Uday V.

PY - 2012

Y1 - 2012

N2 - We consider a class of games, termed as aggregative games, being played over a distributed multi-agent networked system. In an aggregative game, an agent's objective function is coupled through a function of the aggregate of all agents decisions. Every agent maintains an estimate of the aggregate and agents exchange this information over a connected network. We study the gossip-based distributed algorithm for the exchange of information and computation of equilibrium decisions of agents over the network. Our primary emphasis lies in proving the convergence of the algorithm under an assumption of a diminishing (agent-specific) step-size sequence. Under standard conditions, we establish the almost-sure convergence of the generated sequence to the unique equilibrium point.

AB - We consider a class of games, termed as aggregative games, being played over a distributed multi-agent networked system. In an aggregative game, an agent's objective function is coupled through a function of the aggregate of all agents decisions. Every agent maintains an estimate of the aggregate and agents exchange this information over a connected network. We study the gossip-based distributed algorithm for the exchange of information and computation of equilibrium decisions of agents over the network. Our primary emphasis lies in proving the convergence of the algorithm under an assumption of a diminishing (agent-specific) step-size sequence. Under standard conditions, we establish the almost-sure convergence of the generated sequence to the unique equilibrium point.

UR - http://www.scopus.com/inward/record.url?scp=84874277954&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874277954&partnerID=8YFLogxK

U2 - 10.1109/CDC.2012.6426136

DO - 10.1109/CDC.2012.6426136

M3 - Article

AN - SCOPUS:84874277954

SP - 4840

EP - 4845

JO - Scanning Electron Microscopy

JF - Scanning Electron Microscopy

SN - 0586-5581

M1 - 6426136

ER -