On Linear Convergence of PI Consensus Algorithm under the Restricted Secant Inequality

This paper considers solving distributed optimization problems in peer-to-peer multi-agent networks. The network is synchronous and connected. By using the proportional-integral (PI) control strategy, various algorithms with fixed stepsize have been developed. Two notable among them are the PI algor...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2024 Tenth Indian Control Conference (ICC) S. 415 - 420
Hauptverfasser: Chakrabarti, Kushal, Baranwal, Mayank
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 09.12.2024
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract This paper considers solving distributed optimization problems in peer-to-peer multi-agent networks. The network is synchronous and connected. By using the proportional-integral (PI) control strategy, various algorithms with fixed stepsize have been developed. Two notable among them are the PI algorithm and the PI consensus algorithm. Although the PI algorithm has provable linear or exponential convergence without the standard requirement of (strong) convexity, a similar guarantee for the PI consensus algorithm is unavailable. In this paper, using Lyapunov theory, we guarantee exponential convergence of the PI consensus algorithm for global cost functions that satisfy the restricted secant inequality, with rate-matching discretization, without requiring convexity. To accelerate the PI consensus algorithm, we incorporate local pre-conditioning in the form of constant positive definite matrices and numerically validate its efficiency compared to the prominent distributed convex optimization algorithms. Unlike classical pre-conditioning, where only the gradients are multiplied by a pre-conditioner, the proposed pre-conditioning modifies both the gradients and the consensus terms, thereby controlling the effect of the communication graph on the algorithm.
AbstractList This paper considers solving distributed optimization problems in peer-to-peer multi-agent networks. The network is synchronous and connected. By using the proportional-integral (PI) control strategy, various algorithms with fixed stepsize have been developed. Two notable among them are the PI algorithm and the PI consensus algorithm. Although the PI algorithm has provable linear or exponential convergence without the standard requirement of (strong) convexity, a similar guarantee for the PI consensus algorithm is unavailable. In this paper, using Lyapunov theory, we guarantee exponential convergence of the PI consensus algorithm for global cost functions that satisfy the restricted secant inequality, with rate-matching discretization, without requiring convexity. To accelerate the PI consensus algorithm, we incorporate local pre-conditioning in the form of constant positive definite matrices and numerically validate its efficiency compared to the prominent distributed convex optimization algorithms. Unlike classical pre-conditioning, where only the gradients are multiplied by a pre-conditioner, the proposed pre-conditioning modifies both the gradients and the consensus terms, thereby controlling the effect of the communication graph on the algorithm.
Author Chakrabarti, Kushal
Baranwal, Mayank
Author_xml – sequence: 1
  givenname: Kushal
  surname: Chakrabarti
  fullname: Chakrabarti, Kushal
  email: chakrabarti.k@tcs.com
  organization: Tata Consultancy Services Research,Division of Data & Decision Sciences,Mumbai,India,400607
– sequence: 2
  givenname: Mayank
  surname: Baranwal
  fullname: Baranwal, Mayank
  email: baranwal.mayank@tcs.com
  organization: Tata Consultancy Services Research,Division of Data & Decision Sciences,Mumbai,India,400607
BookMark eNo1j8tKw0AYRkfQhda-gci8QOpM5r4swUsgUNHitszlTzuQTnQyEfr2Kurqg7M4nO8KnacxAUK3lKwoJeaubRrJlWCrmtR8RYnWTAlyhpZGGc0YFVTVtL5Eb5uEu5jAZtyM6RPyHpIHPPb4uf0hE6RpnvB62I85lsMRzylAxuUA-AWmkqMvEPAreJsKbhN8zHaI5XSNLno7TLD82wXaPtxvm6eq2zy2zbqroqGlAm7Jd08wJEjrmVeip04bbaUm3FsrHWVS1Vy4oB0N0kkblOTEBU-MFoQt0M2vNgLA7j3Ho82n3f9b9gUmuE9A
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/ICC64753.2024.10883750
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE/IET Electronic Library (IEL) (UW System Shared)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9798331517212
EndPage 420
ExternalDocumentID 10883750
Genre orig-research
GroupedDBID 6IE
6IL
CBEJK
RIE
RIL
ID FETCH-LOGICAL-i91t-e4a0979d90d6ac3c75f1b898a6804caa6b1367245bd8b1d6b6ad7640bdc098503
IEDL.DBID RIE
IngestDate Wed Feb 26 09:42:56 EST 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i91t-e4a0979d90d6ac3c75f1b898a6804caa6b1367245bd8b1d6b6ad7640bdc098503
PageCount 6
ParticipantIDs ieee_primary_10883750
PublicationCentury 2000
PublicationDate 2024-Dec.-9
PublicationDateYYYYMMDD 2024-12-09
PublicationDate_xml – month: 12
  year: 2024
  text: 2024-Dec.-9
  day: 09
PublicationDecade 2020
PublicationTitle 2024 Tenth Indian Control Conference (ICC)
PublicationTitleAbbrev ICC
PublicationYear 2024
Publisher IEEE
Publisher_xml – name: IEEE
Score 1.8918692
Snippet This paper considers solving distributed optimization problems in peer-to-peer multi-agent networks. The network is synchronous and connected. By using the...
SourceID ieee
SourceType Publisher
StartPage 415
SubjectTerms Accelerated aging
Agents-based systems
Consensus algorithm
Convergence
Convex functions
Cost function
Distributed optimization algorithms
Linear matrix inequalities
Logistic regression
Lyapunov methods
Peer-to-peer computing
Title On Linear Convergence of PI Consensus Algorithm under the Restricted Secant Inequality
URI https://ieeexplore.ieee.org/document/10883750
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEA5aPHhSseKbHLxuzW7TPI6yWCxIXbSU3koes1rQXelD8N87ybaKBw_ekhAIzGQy8yXzZQi5UiK16AdNgoC2l3CtIDGsdNi1IJXLjI-5OeN7ORyqyUQXa7J65MIAQEw-g05oxrd8X7tVuCpDC1eIpwJC35ZSNGStNes3Zfp6kOeCY_iNqC_jnc3kX2VTotfo7_1zvX3S_uHf0eLbsxyQLagOyfihoogccWfSPKSKR9Yk0LqkxSCMLELRigW9eX2uEfC_vNFAD5tTDPDoI4TqHA6DS_oEQZZ0UEFDp_xsk1H_dpTfJeuqCMlMp8sEuGFaaq-ZF8Z1neyVqVVaGaEYd8YIGz5hy3jPemVTL6wwXgrOrHcM1cG6R6RV1RUcEyqUAK6cQbvs4nlZal06VClgyFTaTKYnpB1kMn1v_r2YbsRx-sf4GdkNko_JHvqctJbzFVyQHfexnC3ml1FbXzh4l9c
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEA5SBT2pWPFtDl63ZrfZbHKUxdJirUVL6a3kMasF3ZU-BP-9k7RVPHjwloRAYCaTmS-ZL0PIlRSxQT-oIwS0acSVhEizwmLXQCZtol3IzRl2s15PjkaqvyKrBy4MAITkM2j4ZnjLd5Vd-KsytHCJeMoj9M2U84Qt6Vor3m_M1HUnzwXHABxxX8Ib6-m_CqcEv9Ha_eeKe6T-w8Cj_W_fsk82oDwgw4eSInbEvUlznyweeJNAq4L2O35k5stWzOjN63OFkP_ljXqC2JRiiEcfwdfnsBhe0ifw0qSdEpaEys86GbRuB3k7WtVFiCYqnkfANVOZcoo5oW3TZmkRG6mkFpJxq7Uw_hu2hKfGSRM7YYR2meDMOMtQIax5SGplVcIRoUIK4NJqtMwmnpiFUoVFpQIGTYVJsviY1L1Mxu_Lny_Ga3Gc_DF-Sbbbg_vuuNvp3Z2SHa-FkPqhzkhtPl3AOdmyH_PJbHoRNPcFTVybHg
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2024+Tenth+Indian+Control+Conference+%28ICC%29&rft.atitle=On+Linear+Convergence+of+PI+Consensus+Algorithm+under+the+Restricted+Secant+Inequality&rft.au=Chakrabarti%2C+Kushal&rft.au=Baranwal%2C+Mayank&rft.date=2024-12-09&rft.pub=IEEE&rft.spage=415&rft.epage=420&rft_id=info:doi/10.1109%2FICC64753.2024.10883750&rft.externalDocID=10883750