Distributed Asynchronous Deterministic and Stochastic Gradient Optimization Algorithms

John N. Tsitsiklis; Dimitri P. Bertsekas; Michael Athans

doi:10.1109/TAC.1986.1104412

Distributed Asynchronous Deterministic and Stochastic Gradient Optimization Algorithms

John N. Tsitsiklis, Dimitri P. Bertsekas, Michael Athans

Research output: Contribution to journal › Article › peer-review

1393 Scopus citations

Abstract

We present a model for asynchronous distributed computation and then proceed to analyze the convergence of natural asynchronous distributed versions of a large class of deterministic and stochastic gradient-like algorithms. We show that such algorithms retain the desirable convergence properties of their centralized counterparts, provided that the time between consecutive interprocessor communications and the communication delays are not too large.

Original language	English (US)
Pages (from-to)	803-812
Number of pages	10
Journal	IEEE Transactions on Automatic Control
Volume	31
Issue number	9
DOIs	https://doi.org/10.1109/TAC.1986.1104412
State	Published - Sep 1986
Externally published	Yes

ASJC Scopus subject areas

Control and Systems Engineering
Computer Science Applications
Electrical and Electronic Engineering

Access to Document

10.1109/TAC.1986.1104412

Cite this

@article{afe727cd65e44d8ebb1d1e1ad602b087,

title = "Distributed Asynchronous Deterministic and Stochastic Gradient Optimization Algorithms",

abstract = "We present a model for asynchronous distributed computation and then proceed to analyze the convergence of natural asynchronous distributed versions of a large class of deterministic and stochastic gradient-like algorithms. We show that such algorithms retain the desirable convergence properties of their centralized counterparts, provided that the time between consecutive interprocessor communications and the communication delays are not too large.",

author = "Tsitsiklis, {John N.} and Bertsekas, {Dimitri P.} and Michael Athans",

year = "1986",

month = sep,

doi = "10.1109/TAC.1986.1104412",

language = "English (US)",

volume = "31",

pages = "803--812",

journal = "IEEE Transactions on Automatic Control",

issn = "0018-9286",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "9",

}

TY - JOUR

T1 - Distributed Asynchronous Deterministic and Stochastic Gradient Optimization Algorithms

AU - Tsitsiklis, John N.

AU - Bertsekas, Dimitri P.

AU - Athans, Michael

PY - 1986/9

Y1 - 1986/9

N2 - We present a model for asynchronous distributed computation and then proceed to analyze the convergence of natural asynchronous distributed versions of a large class of deterministic and stochastic gradient-like algorithms. We show that such algorithms retain the desirable convergence properties of their centralized counterparts, provided that the time between consecutive interprocessor communications and the communication delays are not too large.

AB - We present a model for asynchronous distributed computation and then proceed to analyze the convergence of natural asynchronous distributed versions of a large class of deterministic and stochastic gradient-like algorithms. We show that such algorithms retain the desirable convergence properties of their centralized counterparts, provided that the time between consecutive interprocessor communications and the communication delays are not too large.

UR - http://www.scopus.com/inward/record.url?scp=0022783899&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0022783899&partnerID=8YFLogxK

U2 - 10.1109/TAC.1986.1104412

DO - 10.1109/TAC.1986.1104412

M3 - Article

AN - SCOPUS:0022783899

SN - 0018-9286

VL - 31

SP - 803

EP - 812

JO - IEEE Transactions on Automatic Control

JF - IEEE Transactions on Automatic Control

IS - 9

ER -

Distributed Asynchronous Deterministic and Stochastic Gradient Optimization Algorithms

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this