A Distributed, Asynchronous, and Incremental Algorithm for Nonconvex Optimization: An ADMM Approach

The alternating direction method of multipliers (ADMM) has been popular for solving many signal processing problems, convex or nonconvex. In this paper, we study an asynchronous implementation of ADMM for solving a nonconvex nonsmooth optimization problem, whose objective is the sum of a number of c...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on control of network systems Jg. 5; H. 3; S. 935 - 945
1. Verfasser: Hong, Mingyi
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Piscataway IEEE 01.09.2018
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
ISSN:2325-5870, 2372-2533
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract The alternating direction method of multipliers (ADMM) has been popular for solving many signal processing problems, convex or nonconvex. In this paper, we study an asynchronous implementation of ADMM for solving a nonconvex nonsmooth optimization problem, whose objective is the sum of a number of component functions. The proposed algorithm allows the problem to be solved in a distributed, asynchronous, and incremental manner. First, the component functions can be distributed to different computing nodes, which perform the updates asynchronously without coordinating with each other. Two sources of asynchrony are covered by our algorithm: One is caused by the heterogeneity of the computational nodes and the other arises from unreliable communication links. Second, the algorithm can be viewed as implementing an incremental algorithm where at each step the (possibly delayed) gradients of only a subset of component functions are updated. We show that when certain bounds are imposed on the level of asynchrony, the proposed algorithm converges to the set of stationary solutions (resp. optimal solutions) for the nonconvex (resp. convex) problem, with a global sublinear rate.
AbstractList The alternating direction method of multipliers (ADMM) has been popular for solving many signal processing problems, convex or nonconvex. In this paper, we study an asynchronous implementation of ADMM for solving a nonconvex nonsmooth optimization problem, whose objective is the sum of a number of component functions. The proposed algorithm allows the problem to be solved in a distributed, asynchronous, and incremental manner. First, the component functions can be distributed to different computing nodes, which perform the updates asynchronously without coordinating with each other. Two sources of asynchrony are covered by our algorithm: One is caused by the heterogeneity of the computational nodes and the other arises from unreliable communication links. Second, the algorithm can be viewed as implementing an incremental algorithm where at each step the (possibly delayed) gradients of only a subset of component functions are updated. We show that when certain bounds are imposed on the level of asynchrony, the proposed algorithm converges to the set of stationary solutions (resp. optimal solutions) for the nonconvex (resp. convex) problem, with a global sublinear rate.
Author Hong, Mingyi
Author_xml – sequence: 1
  givenname: Mingyi
  orcidid: 0000-0003-1263-9365
  surname: Hong
  fullname: Hong, Mingyi
  email: mhong@umn.edu
  organization: Department of Industrial and Manufacturing Systems Engineering, Iowa State University, Ames, IA, USA
BookMark eNp9UMlOwzAQtRBIQOkHIC6WuJJie5o44Ra1bBLLgXKOHGdKjVq72C6ifD0uRRw4cJoZzVv03iHZtc4iIcecDThn1flk9PA0EIzLgShyOSzYDjkQIEUmcoDdzS7yLC8l2yf9EF4ZY1zk6YYDoms6NiF6064idme0DmurZ95ZtwpnVNmO3lrtcYE2qjmt5y_Omzhb0Knz9MFZ7ew7ftDHZTQL86micfaC1pbW4_t7Wi-X3ik9OyJ7UzUP2P-ZPfJ8dTkZ3WR3j9e3o_ou06KCmAHrFLSlaCvsCq1lWQFr9bBqUUORBgdgefoDcFRtm6MY6hZhiKLMC2Qd9MjpVjfZvq0wxObVrbxNlo3gXHKZVCCh-BalvQvB47RZerNQft1w1mzqbDZ1Nps6m586E0f-4WgTv9NGr8z8X-bJlmkQ8ddJlikxCPgCA-CEMw
CODEN ITCNAY
CitedBy_id crossref_primary_10_1109_TSP_2020_2984895
crossref_primary_10_1145_3530693
crossref_primary_10_1109_TSP_2023_3239799
crossref_primary_10_1109_TCNS_2022_3186653
crossref_primary_10_1007_s11227_019_02784_y
crossref_primary_10_1109_TNNLS_2021_3125774
crossref_primary_10_1109_ACCESS_2021_3061995
crossref_primary_10_1109_TNET_2024_3350198
crossref_primary_10_1109_LCSYS_2023_3341100
crossref_primary_10_1016_j_engappai_2025_110626
crossref_primary_10_1137_16M1057000
crossref_primary_10_1109_TAC_2020_3033490
crossref_primary_10_1109_TASE_2020_2991042
crossref_primary_10_1007_s11431_024_2852_1
crossref_primary_10_1109_TSP_2025_3579628
crossref_primary_10_1007_s10107_019_01408_w
crossref_primary_10_3390_en17051233
crossref_primary_10_1137_22M1490995
crossref_primary_10_1109_TSIPN_2020_2994008
crossref_primary_10_1109_TCNS_2025_3526555
crossref_primary_10_1007_s11227_020_03590_7
crossref_primary_10_1016_j_sigpro_2020_107529
crossref_primary_10_1109_TAC_2023_3261465
crossref_primary_10_1109_TCNS_2024_3354840
crossref_primary_10_1109_TPWRS_2022_3162329
crossref_primary_10_1007_s40998_021_00461_6
crossref_primary_10_1186_s13660_019_2145_0
crossref_primary_10_1287_opre_2022_0228
crossref_primary_10_1109_TCNS_2024_3462519
crossref_primary_10_1109_JIOT_2021_3049384
crossref_primary_10_1002_rnc_6048
crossref_primary_10_1109_TASE_2024_3435073
crossref_primary_10_1109_TCNS_2022_3213706
crossref_primary_10_1007_s10589_018_0034_y
crossref_primary_10_1109_TSP_2023_3301621
Cites_doi 10.1016/0898-1221(76)90003-1
10.1109/GlobalSIP.2013.6736937
10.1515/9781400873173
10.1561/2200000016
10.1016/S1570-579X(01)80023-9
10.1109/TPDS.2017.2658620
10.1109/TAC.1986.1104412
10.1109/JSAC.2014.2328171
10.1137/140957639
10.1109/TWC.2010.06.090890
10.1111/j.1467-9868.2011.00783.x
10.1137/140998135
10.1109/TIT.2008.924723
10.1137/110836936
10.1007/BF01581204
10.1109/TSP.2016.2537261
10.1109/TSP.2016.2537271
10.1137/110849468
10.1137/15M1024950
10.1007/s10589-016-9828-y
10.1080/10556788.2012.700713
10.1109/JSAC.2002.1007390
10.1137/140990309
10.1007/s10957-013-0409-2
10.1137/070703983
10.1109/ICASSP.2015.7178689
10.1109/TSP.2015.2454476
10.1007/s10915-015-0048-x
10.1109/TSP.2016.2544743
ContentType Journal Article
Copyright Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2018
Copyright_xml – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2018
DBID 97E
RIA
RIE
AAYXX
CITATION
7SC
8FD
JQ2
L7M
L~C
L~D
DOI 10.1109/TCNS.2017.2657460
DatabaseName IEEE Xplore (IEEE)
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
Computer and Information Systems Abstracts
Technology Research Database
ProQuest Computer Science Collection
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts – Academic
Computer and Information Systems Abstracts Professional
DatabaseTitle CrossRef
Computer and Information Systems Abstracts
Technology Research Database
Computer and Information Systems Abstracts – Academic
Advanced Technologies Database with Aerospace
ProQuest Computer Science Collection
Computer and Information Systems Abstracts Professional
DatabaseTitleList
Computer and Information Systems Abstracts
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 2372-2533
EndPage 945
ExternalDocumentID 10_1109_TCNS_2017_2657460
7829332
Genre orig-research
GrantInformation_xml – fundername: National Science Foundation
  grantid: CCF-1526078
  funderid: 10.13039/501100008982
– fundername: Air Force Office of Scientific Research
  grantid: 15RT0767
  funderid: 10.13039/100000181
GroupedDBID 0R~
4.4
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABJNI
ABQJQ
ABVLG
AGQYO
AGSQL
AHBIQ
AKJIK
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
EBS
EJD
IFIPE
IPLJI
JAVBF
M43
OCL
PQQKQ
RIA
RIE
AAYXX
CITATION
7SC
8FD
JQ2
L7M
L~C
L~D
ID FETCH-LOGICAL-c293t-30da3b82b9ed6cc78930bc49bec3649b13305b82331eabb5e24cbe34e2856e0d3
IEDL.DBID RIE
ISICitedReferencesCount 46
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000445357100011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2325-5870
IngestDate Sun Jun 29 16:08:02 EDT 2025
Sat Nov 29 06:13:46 EST 2025
Tue Nov 18 20:45:15 EST 2025
Wed Aug 27 08:31:58 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 3
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
https://doi.org/10.15223/policy-029
https://doi.org/10.15223/policy-037
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c293t-30da3b82b9ed6cc78930bc49bec3649b13305b82331eabb5e24cbe34e2856e0d3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ORCID 0000-0003-1263-9365
PQID 2117176493
PQPubID 2040410
PageCount 11
ParticipantIDs proquest_journals_2117176493
crossref_citationtrail_10_1109_TCNS_2017_2657460
crossref_primary_10_1109_TCNS_2017_2657460
ieee_primary_7829332
PublicationCentury 2000
PublicationDate 2018-09-01
PublicationDateYYYYMMDD 2018-09-01
PublicationDate_xml – month: 09
  year: 2018
  text: 2018-09-01
  day: 01
PublicationDecade 2010
PublicationPlace Piscataway
PublicationPlace_xml – name: Piscataway
PublicationTitle IEEE transactions on control of network systems
PublicationTitleAbbrev TCNS
PublicationYear 2018
Publisher IEEE
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml – name: IEEE
– name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References razaviyayn (ref50) 2014
ref14
hong (ref22) 2016
ref17
defazio (ref36) 2014
ref16
ref19
ref18
schmidt (ref40) 2013
agarwal (ref13) 2011
zhang (ref32) 2014
bertsekas (ref1) 1999
ref45
glowinski (ref15) 1975; 9
scutari (ref6) 2007
ref47
ref42
ref41
ref44
razaviyayn (ref38) 2013
ref49
hong (ref43) 0
ref7
zhang (ref26) 2010
(ref8) 2013
liu (ref12) 2014
ref4
ref3
ref5
ref35
ref34
richtárik (ref46) 2012
ref31
ref30
ref2
ref39
li (ref10) 2013
jiang (ref48) 2014
ho (ref9) 2013
ref24
ref23
sra (ref37) 2012
ref25
ref20
ref21
ref28
ref27
ref29
niu (ref11) 2011
wei (ref33) 2013
References_xml – ident: ref16
  doi: 10.1016/0898-1221(76)90003-1
– year: 2013
  ident: ref33
  article-title: On the O(1/k) convergence of asynchronous distributed alternating direction method of multipliers
  doi: 10.1109/GlobalSIP.2013.6736937
– start-page: 345
  year: 2007
  ident: ref6
  article-title: Asynchronous iterative water-filling for gaussian frequency-selective interference channels: A unified framework
  publication-title: Proc Inf Theory Appl Workshop
– year: 2013
  ident: ref38
  article-title: A stochastic successive minimization method for nonsmooth nonconvex optimization with applications to transceiver design in wireless communication networks
– ident: ref39
  doi: 10.1515/9781400873173
– volume: 9
  start-page: 41
  year: 1975
  ident: ref15
  article-title: Sur l'approximation, par elements finis d'ordre un,et la resolution, par penalisation-dualite, d'une classe de problemes de dirichlet non lineares
  publication-title: Revue Franqaise d'Automatique Informatique et Rech Opirationelle
– year: 1999
  ident: ref1
  publication-title: Parallel and Distributed Computation Numerical Methods
– ident: ref35
  doi: 10.1561/2200000016
– start-page: 1701
  year: 2014
  ident: ref32
  article-title: Asynchronous distributed ADMM for consensus optimization
  publication-title: Proc 31st Int Conf Mach Learn
– ident: ref4
  doi: 10.1016/S1570-579X(01)80023-9
– year: 2013
  ident: ref8
  publication-title: Frontiers in Massive Data Analysis
– year: 2016
  ident: ref22
  article-title: On the linear convergence of the alternating direction method of multipliers
  publication-title: Math Program Ser A
– ident: ref24
  doi: 10.1109/TPDS.2017.2658620
– ident: ref2
  doi: 10.1109/TAC.1986.1104412
– ident: ref25
  doi: 10.1109/JSAC.2014.2328171
– ident: ref41
  doi: 10.1137/140957639
– year: 2013
  ident: ref40
  article-title: Minimizing finite sums with the stochastic average gradient
  publication-title: INRIA Paris France Tech Rep HAL 00860051
– ident: ref3
  doi: 10.1109/TWC.2010.06.090890
– year: 2014
  ident: ref48
  article-title: Iteration bounds for finding the $\epsilon$ -stationary points for structured nonconvex optimization
– start-page: 1646
  year: 2014
  ident: ref36
  article-title: Saga: A fast incremental gradient method with support for non-strongly convex composite objectives
  publication-title: Proc Adv Neural Inf Process Syst
– start-page: 693
  year: 2011
  ident: ref11
  article-title: HOGWILD!: A lock-free approach to parallelizing stochastic gradient descent
  publication-title: Proc 24th Int Conf Neural Inf Process Syst
– ident: ref49
  doi: 10.1111/j.1467-9868.2011.00783.x
– ident: ref18
  doi: 10.1561/2200000016
– ident: ref31
  doi: 10.1137/140998135
– start-page: 873
  year: 2011
  ident: ref13
  article-title: Distributed delayed stochastic optimization
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref5
  doi: 10.1109/TIT.2008.924723
– ident: ref20
  doi: 10.1137/110836936
– ident: ref17
  doi: 10.1007/BF01581204
– year: 0
  ident: ref43
  article-title: Supplemental material for a distributed, asynchronous and incremental algorithm for nonconvex optimization: An ADMM based approach
– year: 2013
  ident: ref10
  article-title: Distributed delayed proximal gradient methods
  publication-title: Proc NIPS Workshop Optim Mach Learn
– year: 2012
  ident: ref37
  publication-title: Optimization for Machine Learning
– start-page: 1440
  year: 2014
  ident: ref50
  article-title: Parallel successive convex approximation for nonsmooth nonconvex optimization
  publication-title: Proc Neural Inf Process Syst
– start-page: 469
  year: 2014
  ident: ref12
  article-title: An asynchronous parallel stochastic coordinate descent algorithm
  publication-title: Proc Int Conf Mach Learn
– ident: ref45
  doi: 10.1109/TSP.2016.2537261
– year: 2012
  ident: ref46
  article-title: Alternating maximization: Unifying framework for 8 Sparse PCA formulations and efficient parallel codes
– ident: ref44
  doi: 10.1109/TSP.2016.2537271
– ident: ref21
  doi: 10.1137/110849468
– ident: ref14
  doi: 10.1137/15M1024950
– ident: ref27
  doi: 10.1007/s10589-016-9828-y
– ident: ref29
  doi: 10.1080/10556788.2012.700713
– ident: ref7
  doi: 10.1109/JSAC.2002.1007390
– ident: ref30
  doi: 10.1137/140990309
– ident: ref42
  doi: 10.1007/s10957-013-0409-2
– start-page: 10
  year: 2010
  ident: ref26
  article-title: An alternating direction algorithm for nonnegative matrix factorization
  publication-title: Rice Univ Houston TX USA Tech Rep TR
– ident: ref23
  doi: 10.1137/070703983
– ident: ref47
  doi: 10.1109/ICASSP.2015.7178689
– ident: ref28
  doi: 10.1109/TSP.2015.2454476
– ident: ref19
  doi: 10.1007/s10915-015-0048-x
– ident: ref34
  doi: 10.1109/TSP.2016.2544743
– start-page: 1223
  year: 2013
  ident: ref9
  article-title: More effective distributed ML via a stale synchronous parallel parameter server
  publication-title: Proc Adv Neural Inf Process Syst
SSID ssj0001255873
Score 2.3896616
Snippet The alternating direction method of multipliers (ADMM) has been popular for solving many signal processing problems, convex or nonconvex. In this paper, we...
SourceID proquest
crossref
ieee
SourceType Aggregation Database
Enrichment Source
Index Database
Publisher
StartPage 935
SubjectTerms Algorithm design and analysis
Algorithms
Alternating direction method of multipliers
asynchronous optimization
Control systems
Delays
distributed optimization
Machine learning algorithms
Nodes
Optimization
Signal processing
Signal processing algorithms
Title A Distributed, Asynchronous, and Incremental Algorithm for Nonconvex Optimization: An ADMM Approach
URI https://ieeexplore.ieee.org/document/7829332
https://www.proquest.com/docview/2117176493
Volume 5
WOSCitedRecordID wos000445357100011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVIEE
  databaseName: IEEE Electronic Library (IEL)
  customDbUrl:
  eissn: 2372-2533
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001255873
  issn: 2325-5870
  databaseCode: RIE
  dateStart: 20140101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NT8IwFG-UeNCDX2hE0fTgyTAY67Zu3haQeBA0EQ23ZWsfSgLDwDD63_vaFSTRmHjZlqxtuv22996vfR-EXIqUS2RcroXGx9ByIUislOFBwtBxk2AYSr17_nzHe71gMAgfNkhtFQsDANr5DOrqUu_ly6lYqKWyBmoz5N8ocDc594tYrbX1FM8LODMbl007bPRbvUflu8Xrju9xVyeh_FY9upbKDwGstUpn73_z2Se7xnqkUQH3AdmA7JDsrOUULBMR0bZKhqvqWIGs0Wj-mQmVARcpfo0mmaQoEopFQTXS-GU6G-WvE4rGK-1NM-2F_kHvUZJMTIjmNY0yGrW7XRqZBORH5Klz02_dWqaSgiVwhrnFbJmwNHDSEKQvBEcjxU6FGyKAzMcTElXbw_uMNSFJUw8cV6TAXHACzwdbsmNSyqYZnBCKKh45SNNJhqj8E4-HHk84hMJXik7askLs5UuOhUkzrqpdjGNNN-wwVrjECpfY4FIhV6sub0WOjb8alxUQq4YGgwqpLpGMzV84j5HcIlvFB2Snv_c6I9s4dlD4jFVJKZ8t4Jxsifd8NJ9d6A_sCwsezwk
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1RT9swED6hgjR4gI2CKCvDD3tCDU3jJI73FhVQ0doMad3EW5TY1w0JUtQWBP-es2MKEtMkXpJIsRMnX3J33_l8B_BVlUIT4wo9Mj4mXohJ4ZWcNhonQVgkE6nt7Pnvociy5PJSXqxAZ7kWBhFt8Bkem0M7l6-n6s64yrqkzYh_k8BdNZWzZL1a65VHJYoSwd3UZc-X3XE_-2mit8RxEEcitGkoX5SPrabyRgRbvXK29b4RfYRNZz-ytAb8E6xgtQ0br7IKNkGl7MSkwzWVrFB3WDp_rJTJgUskv8OKSjMSCrVb0Fzp-s90drX4e8PIfGXZtLJx6A_sB8mSG7dI8xtLK5aejEYsdSnId-DX2em4P_BcLQVP0QgXHvd1wcskKCXqWClBZopfqlAShDymHVFVP6LznPewKMsIg1CVyEMMkihGX_NdaFTTCveAkZInFtILigmp_yISMhKFQKlio-q0r1vgP7_kXLlE46bexXVuCYcvc4NLbnDJHS4tOFp2ua2zbPyvcdMAsWzoMGhB-xnJ3P2H85zoLfFVekC-_-9eh_BhMB4N8-F59v0zrNN9kjqCrA2NxewOD2BN3S-u5rMv9mN7Agv10lg
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%3Ajournal&rft.genre=article&rft.atitle=A+Distributed%2C+Asynchronous%2C+and+Incremental+Algorithm+for+Nonconvex+Optimization%3A+An+ADMM+Approach&rft.jtitle=IEEE+transactions+on+control+of+network+systems&rft.au=Hong%2C+Mingyi&rft.date=2018-09-01&rft.pub=IEEE&rft.eissn=2372-2533&rft.volume=5&rft.issue=3&rft.spage=935&rft.epage=945&rft_id=info:doi/10.1109%2FTCNS.2017.2657460&rft.externalDocID=7829332
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2325-5870&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2325-5870&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2325-5870&client=summon