Stochastic Proximal Gradient Consensus Over Random Networks

We consider solving a convex optimization problem with possibly stochastic gradient, and over a randomly time-varying multiagent network. Each agent has access to some local objective function, and it only has unbiased estimates of the gradients of the smooth component. We develop a dynamic stochast...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on signal processing Jg. 65; H. 11; S. 2933 - 2948
Hauptverfasser: Hong, Mingyi, Chang, Tsung-Hui
Format: Journal Article
Sprache:Englisch
Veröffentlicht: IEEE 01.06.2017
Schlagworte:
ISSN:1053-587X, 1941-0476
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract We consider solving a convex optimization problem with possibly stochastic gradient, and over a randomly time-varying multiagent network. Each agent has access to some local objective function, and it only has unbiased estimates of the gradients of the smooth component. We develop a dynamic stochastic proximal-gradient consensus algorithm, with the following key features: (1) it works for both the static and certain randomly time-varying networks; (2) it allows the agents to utilize either the exact or stochastic gradient information; (3) it is convergent with provable rate. In particular, the proposed algorithm converges to a global optimal solution, with a rate of O(1/r) [resp. O(1/√r)] when the exact (resp. stochastic) gradient is available, where r is the iteration counter. Interestingly, the developed algorithm establishes a close connection among a number of (seemingly unrelated) distributed algorithms, such as the EXTRA, the PG-EXTRA, the IC/IDC-ADMM, the DLM, and the classical distributed subgradient method.
AbstractList We consider solving a convex optimization problem with possibly stochastic gradient, and over a randomly time-varying multiagent network. Each agent has access to some local objective function, and it only has unbiased estimates of the gradients of the smooth component. We develop a dynamic stochastic proximal-gradient consensus algorithm, with the following key features: (1) it works for both the static and certain randomly time-varying networks; (2) it allows the agents to utilize either the exact or stochastic gradient information; (3) it is convergent with provable rate. In particular, the proposed algorithm converges to a global optimal solution, with a rate of O(1/r) [resp. O(1/√r)] when the exact (resp. stochastic) gradient is available, where r is the iteration counter. Interestingly, the developed algorithm establishes a close connection among a number of (seemingly unrelated) distributed algorithms, such as the EXTRA, the PG-EXTRA, the IC/IDC-ADMM, the DLM, and the classical distributed subgradient method.
Author Tsung-Hui Chang
Mingyi Hong
Author_xml – sequence: 1
  givenname: Mingyi
  surname: Hong
  fullname: Hong, Mingyi
– sequence: 2
  givenname: Tsung-Hui
  surname: Chang
  fullname: Chang, Tsung-Hui
BookMark eNp9j01LAzEURYNUsK3uBTfzB6bmzeQTV1K0CsUWW8HdkGYyGG0TSaLVf29KiwsXrt7dnPvuGaCe884gdA54BIDl5XIxH1UY-KhivBZAj1AfJIESE856OWNal1Tw5xM0iPEVYyBEsj66WiSvX1RMVhfz4L_sRq2LSVCtNS4VY--icfEjFrNPE4pH5Vq_KR5M2vrwFk_RcafW0Zwd7hA93d4sx3fldDa5H19PS12xOpVS0HYFnYEVobjVglPCuWbY5FArRgkGBR0B2VZKAhfMSIVV3so6w8WK1UOE9706-BiD6Zr3kHeG7wZws5Nvsnyzk28O8hlhfxBtk0rWuxSUXf8HXuxBa4z5_ZNHVUKw-geJAGiP
CODEN ITPRED
CitedBy_id crossref_primary_10_1109_ACCESS_2019_2923269
crossref_primary_10_1109_TAC_2020_2989266
crossref_primary_10_1016_j_jpdc_2022_03_014
crossref_primary_10_1080_00207721_2020_1815098
crossref_primary_10_1109_TAC_2020_2989282
crossref_primary_10_1109_TAC_2021_3092561
crossref_primary_10_1109_JIOT_2025_3534982
crossref_primary_10_1109_TIFS_2021_3050603
crossref_primary_10_1109_ACCESS_2023_3341295
crossref_primary_10_1109_TSMC_2019_2960770
crossref_primary_10_1016_j_sigpro_2020_107614
crossref_primary_10_1109_TSIPN_2017_2695121
crossref_primary_10_1016_j_sigpro_2021_108020
crossref_primary_10_1109_TAC_2024_3525182
crossref_primary_10_1109_TSMC_2019_2963452
crossref_primary_10_1109_TAC_2019_2936355
crossref_primary_10_1109_TAC_2021_3049345
crossref_primary_10_1109_TSP_2021_3097211
crossref_primary_10_1109_JIOT_2024_3376552
crossref_primary_10_1109_TAC_2020_3018743
crossref_primary_10_1109_TAC_2022_3151734
crossref_primary_10_1109_TSMC_2022_3144187
crossref_primary_10_1137_20M138956X
crossref_primary_10_1007_s10957_021_01833_y
crossref_primary_10_1016_j_ifacol_2020_12_1290
crossref_primary_10_1109_TAC_2017_2713046
crossref_primary_10_1109_TAC_2023_3261465
crossref_primary_10_1063_5_0275380
crossref_primary_10_1631_FITEE_1900661
crossref_primary_10_1109_TPDS_2021_3072373
crossref_primary_10_1109_TSMC_2019_2946287
crossref_primary_10_1007_s10957_023_02336_8
crossref_primary_10_3390_math10173135
crossref_primary_10_1109_TSIPN_2019_2892055
crossref_primary_10_1109_JSAC_2021_3078495
crossref_primary_10_1109_TAC_2022_3173171
crossref_primary_10_1080_02331934_2024_2407019
crossref_primary_10_1109_TSP_2019_2926022
crossref_primary_10_1109_TAC_2019_2896025
crossref_primary_10_1109_TICPS_2025_3538690
Cites_doi 10.1109/TSP.2016.2544743
10.1137/140990309
10.1137/080716542
10.1109/TAC.2014.2364096
10.1109/TAC.2017.2713046
10.1109/TAC.2010.2041686
10.1109/TSP.2007.906734
10.1109/TSP.2014.2367458
10.1137/14096668X
10.1109/TAC.2008.2009515
10.1109/TSP.2011.2162831
10.1109/TAC.2011.2161027
10.1016/j.jpdc.2006.08.010
10.1109/JSTSP.2011.2114324
10.1109/TAC.2014.2363299
10.1109/ICASSP.2015.7178514
10.1109/TSP.2015.2436358
10.1109/TSP.2008.919636
10.1109/TSIPN.2016.2613678
10.1109/TSP.2013.2254478
10.1561/2200000016
10.1137/140971233
10.1109/ALLERTON.2014.7028466
10.1109/TAC.2014.2298712
10.1109/ICASSP.2016.7472584
10.1109/JSTSP.2011.2118740
10.1109/TWC.2010.06.090890
10.1007/s10957-010-9737-7
10.1109/TSP.2010.2055862
10.1109/MLSP.2014.6958866
10.1017/CBO9780511804458.011
10.1109/GlobalSIP.2013.6736937
10.1109/CDC.2008.4738860
10.1007/978-1-4419-8853-9
10.1137/14095697X
10.1007/978-3-662-12613-4
10.1109/ICASSP.2014.6855109
10.1109/TSP.2014.2304432
ContentType Journal Article
DBID 97E
RIA
RIE
AAYXX
CITATION
DOI 10.1109/TSP.2017.2673815
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005–Present
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 1941-0476
EndPage 2948
ExternalDocumentID 10_1109_TSP_2017_2673815
7862886
Genre orig-research
GrantInformation_xml – fundername: Air Force Office of Scientific Research
  grantid: 15RT0767
  funderid: 10.13039/100000181
– fundername: National Science Foundation
  grantid: CCF-1526078
  funderid: 10.13039/100000001
– fundername: National Natural Science Foundation of China
  grantid: 61571385
  funderid: 10.13039/501100001809
GroupedDBID -~X
.DC
0R~
29I
4.4
5GY
6IK
85S
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACGFO
ACIWK
ACNCT
AENEX
AGQYO
AHBIQ
AJQPL
AKJIK
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ASUFR
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
EBS
EJD
F5P
HZ~
IFIPE
IPLJI
JAVBF
LAI
MS~
O9-
OCL
P2P
RIA
RIE
RNS
TAE
TN5
AAYXX
CITATION
ID FETCH-LOGICAL-c263t-985db1fe1b450dc875477c60e7543a65401a1f419d2a91786e9a0a5876fe78b63
IEDL.DBID RIE
ISICitedReferencesCount 48
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000399727000013&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1053-587X
IngestDate Sat Nov 29 04:10:42 EST 2025
Tue Nov 18 22:37:44 EST 2025
Tue Aug 26 17:00:18 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 11
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-c263t-985db1fe1b450dc875477c60e7543a65401a1f419d2a91786e9a0a5876fe78b63
PageCount 16
ParticipantIDs ieee_primary_7862886
crossref_citationtrail_10_1109_TSP_2017_2673815
crossref_primary_10_1109_TSP_2017_2673815
PublicationCentury 2000
PublicationDate 2017-June1,-1
2017-6-1
PublicationDateYYYYMMDD 2017-06-01
PublicationDate_xml – month: 06
  year: 2017
  text: 2017-June1,-1
  day: 01
PublicationDecade 2010
PublicationTitle IEEE transactions on signal processing
PublicationTitleAbbrev TSP
PublicationYear 2017
Publisher IEEE
Publisher_xml – name: IEEE
References ref13
ref12
ref15
ref14
ref11
ref10
ref17
ref19
ref18
tsitsiklis (ref2) 1984
hong (ref46) 2015
hajinezhad (ref51) 2016
ref48
ref47
ref42
ref44
ref8
ref7
ref4
ref5
nesterov (ref52) 2004
ref40
mokhtari (ref20) 2016; 17
ref35
ref37
ram (ref38) 2010; 147
ref36
ref31
ref30
ref33
ref32
ref1
le roux (ref49) 0
ref39
bertsekas (ref22) 1999
chen (ref16) 2012
gao (ref41) 2014
ref24
shalev-shwartz (ref6) 2013; 14
ref23
ref26
ref25
ref21
zhong (ref43) 2014
ref28
ouyang (ref45) 2013
ref27
ref29
xiao (ref3) 2007; 67
pesquet (ref34) 2015; 16
johnson (ref50) 2013
giannakis (ref9) 2015
References_xml – ident: ref40
  doi: 10.1109/TSP.2016.2544743
– ident: ref33
  doi: 10.1137/140990309
– ident: ref48
  doi: 10.1137/080716542
– ident: ref19
  doi: 10.1109/TAC.2014.2364096
– start-page: 80
  year: 2013
  ident: ref45
  article-title: Stochastic alternating direction method of multipliers
  publication-title: Proc Int Conf Mach Learn
– ident: ref32
  doi: 10.1109/TAC.2017.2713046
– ident: ref11
  doi: 10.1109/TAC.2010.2041686
– ident: ref7
  doi: 10.1109/TSP.2007.906734
– volume: 14
  start-page: 567
  year: 2013
  ident: ref6
  article-title: Proximal stochastic dual coordinate ascent methods for regularized loss minimization
  publication-title: J Mach Learn Res
– year: 2015
  ident: ref46
  article-title: Stochastic proximal gradient consensus over random networks
– start-page: 2672
  year: 0
  ident: ref49
  article-title: A stochastic gradient method with an exponential convergence rate for strongly-convex optimization with finite training sets
  publication-title: Proc Neural Inf Process Syst
– ident: ref30
  doi: 10.1109/TSP.2014.2367458
– ident: ref14
  doi: 10.1137/14096668X
– year: 1999
  ident: ref22
  publication-title: Parallel and Distributed Computation Numerical Methods
– ident: ref10
  doi: 10.1109/TAC.2008.2009515
– volume: 16
  start-page: 2453
  year: 2015
  ident: ref34
  article-title: A class of randomized primal-dual algorithms for distributed optimization
  publication-title: J Nonlinear Convex Anal
– ident: ref29
  doi: 10.1109/TSP.2011.2162831
– volume: 17
  start-page: 2165
  year: 2016
  ident: ref20
  article-title: DSA: Decentralized double stochastic averaging gradient algorithm
  publication-title: J Mach Learn Res
– ident: ref18
  doi: 10.1109/TAC.2011.2161027
– volume: 67
  start-page: 33
  year: 2007
  ident: ref3
  article-title: Distributed average consensus with least-mean-square deviation
  publication-title: J Parallel Distrib Comput
  doi: 10.1016/j.jpdc.2006.08.010
– start-page: 78
  year: 2014
  ident: ref43
  article-title: Fast stochastic alternating direction method of multipliers
  publication-title: Proc 31st Int Conf Mach Learn
– ident: ref4
  doi: 10.1109/JSTSP.2011.2114324
– ident: ref37
  doi: 10.1109/TAC.2014.2363299
– ident: ref15
  doi: 10.1109/ICASSP.2015.7178514
– ident: ref31
  doi: 10.1109/TSP.2015.2436358
– ident: ref47
  doi: 10.1109/TSP.2008.919636
– ident: ref21
  doi: 10.1109/TSIPN.2016.2613678
– ident: ref28
  doi: 10.1109/TSP.2013.2254478
– ident: ref23
  doi: 10.1561/2200000016
– ident: ref36
  doi: 10.1137/140971233
– ident: ref27
  doi: 10.1109/ALLERTON.2014.7028466
– ident: ref17
  doi: 10.1109/TAC.2014.2298712
– start-page: 3215
  year: 2013
  ident: ref50
  article-title: Accelerating stochastic gradient descent using predictive variance reduction
  publication-title: Proc 26th Int Conf Neural Inf Process
– ident: ref1
  doi: 10.1109/ICASSP.2016.7472584
– ident: ref13
  doi: 10.1109/JSTSP.2011.2118740
– ident: ref8
  doi: 10.1109/TWC.2010.06.090890
– year: 2015
  ident: ref9
  article-title: Proximal splitting methods in signal processing
  publication-title: Splitting Methods in Communication and Imaging
– volume: 147
  start-page: 516
  year: 2010
  ident: ref38
  article-title: Distributed stochastic subgradient projection algorithms for convex optimization
  publication-title: J Optim Theory Appl
  doi: 10.1007/s10957-010-9737-7
– year: 1984
  ident: ref2
  article-title: Problems in decentralized decision making and computation
– year: 2012
  ident: ref16
  article-title: Fast distributed first-order methods
– start-page: 3215
  year: 2016
  ident: ref51
  article-title: NESTT: A nonconvex primal-dual splitting method for distributed and stochastic optimization
  publication-title: Proc Neural Inf Process Syst
– ident: ref5
  doi: 10.1109/TSP.2010.2055862
– ident: ref35
  doi: 10.1109/MLSP.2014.6958866
– ident: ref39
  doi: 10.1017/CBO9780511804458.011
– ident: ref25
  doi: 10.1109/GlobalSIP.2013.6736937
– ident: ref12
  doi: 10.1109/CDC.2008.4738860
– year: 2004
  ident: ref52
  publication-title: Introductory Lectures on Convex Optimization A Basic Course
  doi: 10.1007/978-1-4419-8853-9
– ident: ref42
  doi: 10.1137/14095697X
– ident: ref24
  doi: 10.1007/978-3-662-12613-4
– ident: ref44
  doi: 10.1109/ICASSP.2014.6855109
– ident: ref26
  doi: 10.1109/TSP.2014.2304432
– year: 2014
  ident: ref41
  article-title: On the information-adaptive variants of the ADMM: An iteration complexity perspective
SSID ssj0014496
Score 2.5083704
Snippet We consider solving a convex optimization problem with possibly stochastic gradient, and over a randomly time-varying multiagent network. Each agent has access...
SourceID crossref
ieee
SourceType Enrichment Source
Index Database
Publisher
StartPage 2933
SubjectTerms ADMM
Algorithm design and analysis
Convergence
Convex functions
Distributed optimization
fast algorithms
Heuristic algorithms
Linear programming
Optimization
rate analysis
Signal processing algorithms
Title Stochastic Proximal Gradient Consensus Over Random Networks
URI https://ieeexplore.ieee.org/document/7862886
Volume 65
WOSCitedRecordID wos000399727000013&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: 1941-0476
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0014496
  issn: 1053-587X
  databaseCode: RIE
  dateStart: 19910101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NS8MwFA_b8KAHv6Y4v-jBi2C2pk3zgScRpweZw03YraRJigPXSteJf75J2pUdRPAWwntQfi807-V9_AC4UoGglCUCKvPvg9jIQC5RApVAIdI4NF6-axR-pqMRm834uAVuml4YrbUrPtN9u3S5fJXLlX0qG1BmyXFJG7QpJVWvVpMxwNhxcRl3IYQRo7N1StLng-lkbGu4aD-wFJeWAHfjCtrgVHFXynDvfx-zD3Zr19G7q2x9AFo6OwQ7GwMFu-B2UubyXdjZy964yL_nC6PwWLiyrtKz5JyW2WLpvZgD7L2KTOULb1QVgi-PwNvwYXr_BGt6BCgDEpaQs0glKNUowZGvpAk8MKWS-NosQkGMK4YESjHixh4mKGNEc-ELgw1JtTEPCY9BJ8szfQI8jrgOVCKtRXHKFDM-tyKhL4WMGBGyBwZrxGJZzw63FBYfsYshfB4bjGOLcVxj3APXjcZnNTfjD9muhbeRq5E9_X37DGxb5apc6xx0ymKlL8CW_Crny-LSHYofeIG01A
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1bS8MwFD7MKagP3qY4r33wRbCu6SVN8EnEOXHW4SbsraRJigPXSteJP9-k7coeRPAtlJMSvhOSc3IuH8CFsJnvk4iZQp19pqtkTMpRZAqGHCRdR1n5RaFw3w8CMh7TQQOu6loYKWWRfCav9bCI5YuUz_VTWccnmhwXr8CqZs6qqrXqmIHrFmxcymBwTI_440VQ0qKd0XCgs7j8a1uTXGoK3KVLaIlVpbhUutv_W84ObFXGo3FbansXGjLZg82lloItuBnmKX9nuvuyMcjS78lUTXjIisSu3ND0nJrbYma8qC1svLJEpFMjKFPBZ_vw1r0f3fXMiiDB5DZ2cpMST0QolihyPUtw5Xq4vs-xJdXAYVgZY4ih2EVUaUS5ZQRLyiymsMGxVArCzgE0kzSRh2BQRKUtIq516sZEEGV1C-xYnHGPYMbb0FkgFvKqe7gmsfgICy_CoqHCONQYhxXGbbisZ3yWnTP-kG1peGu5Ctmj3z-fw3pv9NwP-4_B0zFs6B-VyVsn0MyzuTyFNf6VT2bZWbFBfgATxbgd
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=Stochastic+Proximal+Gradient+Consensus+Over+Random+Networks&rft.jtitle=IEEE+transactions+on+signal+processing&rft.au=Hong%2C+Mingyi&rft.au=Chang%2C+Tsung-Hui&rft.date=2017-06-01&rft.issn=1053-587X&rft.eissn=1941-0476&rft.volume=65&rft.issue=11&rft.spage=2933&rft.epage=2948&rft_id=info:doi/10.1109%2FTSP.2017.2673815&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TSP_2017_2673815
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1053-587X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1053-587X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1053-587X&client=summon