Convergence rate of distributed averaging dynamics and optimization in networks

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

Recent advances in wired and wireless technology lead to the emergence of large-scale networks such as Internet, wireless mobile ad-hoc networks, swarm robotics, smart-grid, and smart-sensor networks. The advances gave rise to new applications in networks including decentralized resource allocation in multi-agent systems, decentralized control of multi-agent systems, collaborative decision making, decentralized learning and estimation, and decentralized in-network signal processing. The advances also gave birth to new large cyber-physical systems such as sensor and social networks. These network systems are typically spatially distributed over a large area and may consists of hundreds of agents in smart-sensor networks to millions of agents in social networks. As such, they do not possess a central coordinator or a central point for access to the complete system information. This lack of central entity makes the traditional (centralized) optimization and control techniques inapplicable, thus necessitating the development of new distributed computational models and algorithms to support efficient operations over such networks. This tutorial provides an overview of the convergence rate of distributed algorithms for coordination and its relevance to optimization in a system of autonomous agents embedded in a communication network, where each agent is aware of (and can communicate with) its local neighbors only. The focus is on distributed averaging dynamics for consensus problems and its role in consensusbased gradient methods for convex optimization problems, where the network objective function is separable across the constituent agents.

Original languageEnglish (US)
Pages (from-to)1-100
Number of pages100
JournalFoundations and Trends in Systems and Control
Volume2
Issue number1
DOIs
StatePublished - Jan 1 2015
Externally publishedYes

Fingerprint

Smart sensors
Multi agent systems
Sensor networks
Averaging
Rate of Convergence
Autonomous agents
Decentralized control
Gradient methods
Optimization
Convex optimization
Wireless ad hoc networks
Mobile ad hoc networks
Parallel algorithms
Decentralized
Resource allocation
Telecommunication networks
Sensor Networks
Smart Sensors
Signal processing
Robotics

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Control and Optimization

Cite this

Convergence rate of distributed averaging dynamics and optimization in networks. / Nedich, Angelia.

In: Foundations and Trends in Systems and Control, Vol. 2, No. 1, 01.01.2015, p. 1-100.

Research output: Contribution to journalArticle

@article{547dc6c519594ae2b3ab235cc3c9eaca,
title = "Convergence rate of distributed averaging dynamics and optimization in networks",
abstract = "Recent advances in wired and wireless technology lead to the emergence of large-scale networks such as Internet, wireless mobile ad-hoc networks, swarm robotics, smart-grid, and smart-sensor networks. The advances gave rise to new applications in networks including decentralized resource allocation in multi-agent systems, decentralized control of multi-agent systems, collaborative decision making, decentralized learning and estimation, and decentralized in-network signal processing. The advances also gave birth to new large cyber-physical systems such as sensor and social networks. These network systems are typically spatially distributed over a large area and may consists of hundreds of agents in smart-sensor networks to millions of agents in social networks. As such, they do not possess a central coordinator or a central point for access to the complete system information. This lack of central entity makes the traditional (centralized) optimization and control techniques inapplicable, thus necessitating the development of new distributed computational models and algorithms to support efficient operations over such networks. This tutorial provides an overview of the convergence rate of distributed algorithms for coordination and its relevance to optimization in a system of autonomous agents embedded in a communication network, where each agent is aware of (and can communicate with) its local neighbors only. The focus is on distributed averaging dynamics for consensus problems and its role in consensusbased gradient methods for convex optimization problems, where the network objective function is separable across the constituent agents.",
author = "Angelia Nedich",
year = "2015",
month = "1",
day = "1",
doi = "10.1561/2600000004",
language = "English (US)",
volume = "2",
pages = "1--100",
journal = "Foundations and Trends in Systems and Control",
issn = "2325-6818",
publisher = "Now Publishers Inc",
number = "1",

}

TY - JOUR

T1 - Convergence rate of distributed averaging dynamics and optimization in networks

AU - Nedich, Angelia

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Recent advances in wired and wireless technology lead to the emergence of large-scale networks such as Internet, wireless mobile ad-hoc networks, swarm robotics, smart-grid, and smart-sensor networks. The advances gave rise to new applications in networks including decentralized resource allocation in multi-agent systems, decentralized control of multi-agent systems, collaborative decision making, decentralized learning and estimation, and decentralized in-network signal processing. The advances also gave birth to new large cyber-physical systems such as sensor and social networks. These network systems are typically spatially distributed over a large area and may consists of hundreds of agents in smart-sensor networks to millions of agents in social networks. As such, they do not possess a central coordinator or a central point for access to the complete system information. This lack of central entity makes the traditional (centralized) optimization and control techniques inapplicable, thus necessitating the development of new distributed computational models and algorithms to support efficient operations over such networks. This tutorial provides an overview of the convergence rate of distributed algorithms for coordination and its relevance to optimization in a system of autonomous agents embedded in a communication network, where each agent is aware of (and can communicate with) its local neighbors only. The focus is on distributed averaging dynamics for consensus problems and its role in consensusbased gradient methods for convex optimization problems, where the network objective function is separable across the constituent agents.

AB - Recent advances in wired and wireless technology lead to the emergence of large-scale networks such as Internet, wireless mobile ad-hoc networks, swarm robotics, smart-grid, and smart-sensor networks. The advances gave rise to new applications in networks including decentralized resource allocation in multi-agent systems, decentralized control of multi-agent systems, collaborative decision making, decentralized learning and estimation, and decentralized in-network signal processing. The advances also gave birth to new large cyber-physical systems such as sensor and social networks. These network systems are typically spatially distributed over a large area and may consists of hundreds of agents in smart-sensor networks to millions of agents in social networks. As such, they do not possess a central coordinator or a central point for access to the complete system information. This lack of central entity makes the traditional (centralized) optimization and control techniques inapplicable, thus necessitating the development of new distributed computational models and algorithms to support efficient operations over such networks. This tutorial provides an overview of the convergence rate of distributed algorithms for coordination and its relevance to optimization in a system of autonomous agents embedded in a communication network, where each agent is aware of (and can communicate with) its local neighbors only. The focus is on distributed averaging dynamics for consensus problems and its role in consensusbased gradient methods for convex optimization problems, where the network objective function is separable across the constituent agents.

UR - http://www.scopus.com/inward/record.url?scp=84984816627&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84984816627&partnerID=8YFLogxK

U2 - 10.1561/2600000004

DO - 10.1561/2600000004

M3 - Article

VL - 2

SP - 1

EP - 100

JO - Foundations and Trends in Systems and Control

JF - Foundations and Trends in Systems and Control

SN - 2325-6818

IS - 1

ER -