Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Asynchronous incremental block-coordinate descent
KTH, Skolan för elektro- och systemteknik (EES), Reglerteknik.ORCID-id: 0000-0003-1725-2901
KTH, Skolan för elektro- och systemteknik (EES), Reglerteknik.
KTH, Skolan för elektro- och systemteknik (EES), Reglerteknik.
2014 (engelsk)Inngår i: 52nd Annual Allerton Conference on Communication, Control, and Computing (Allerton), IEEE conference proceedings, 2014, s. 19-24Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

This paper studies a flexible algorithm for minimizing a sum of component functions, each of which depends on a large number of decision variables. Such formulations appear naturally in “big data” applications, where each function describes the loss estimated using the data available at a specific machine, and the number of features under consideration is huge. In our algorithm, a coordinator updates a global iterate based on delayed partial gradients of the individual objective functions with respect to blocks of coordinates. Delayed incremental gradient and delayed coordinate descent algorithms are obtained as special cases. Under the assumption of strong convexity and block coordinate-wise Lipschitz continuous partial gradients, we show that the algorithm converges linearly to a ball around the optimal value. Contrary to related proposals in the literature, our algorithm is delay-insensitive: it converges for any bounded information delay, and its step-size parameter can be chosen independently of the maximum delay bound.

sted, utgiver, år, opplag, sider
IEEE conference proceedings, 2014. s. 19-24
Emneord [en]
Big Data, gradient methods, asynchronous incremental block-coordinate descent, big data applications, block coordinate-wise Lipschitz continuous partial gradients;bounded information delay, component function sum minimization, delayed partial gradients, global iterate, maximum delay bound, step-size parameter, Algorithm design and analysis, Color, Convergence, Delays, Servers, Standards, Vectors
HSV kategori
Identifikatorer
URN: urn:nbn:se:kth:diva-164288DOI: 10.1109/ALLERTON.2014.7028430Scopus ID: 2-s2.0-84946692347OAI: oai:DiVA.org:kth-164288DiVA, id: diva2:805300
Konferanse
52nd Annual Allerton Conference on Communication, Control, and Computing (Allerton),Sept. 30 2014-Oct. 3 2014,Monticello, IL
Merknad

QC 20150511

Tilgjengelig fra: 2015-04-15 Laget: 2015-04-15 Sist oppdatert: 2015-05-11bibliografisk kontrollert

Open Access i DiVA

fulltext(851 kB)175 nedlastinger
Filinformasjon
Fil FULLTEXT01.pdfFilstørrelse 851 kBChecksum SHA-512
2bf4d04444de0e488e81a7464e17e2ec32d552fa209b8b94823a7364be2cde5e05a9f365489c859963245cf93f3d8fa74acd84a8a43ee5673b2c15cc699ee6dc
Type fulltextMimetype application/pdf

Andre lenker

Forlagets fulltekstScopusIEEEXplore

Personposter BETA

Aytekin, Arda

Søk i DiVA

Av forfatter/redaktør
Aytekin, ArdaFeyzmahdavian, Hamid RezaJohansson, Mikael
Av organisasjonen

Søk utenfor DiVA

GoogleGoogle Scholar
Totalt: 175 nedlastinger
Antall nedlastinger er summen av alle nedlastinger av alle fulltekster. Det kan for eksempel være tidligere versjoner som er ikke lenger tilgjengelige

doi
urn-nbn

Altmetric

doi
urn-nbn
Totalt: 205 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf