Linearly Convergent Decentralized Consensus Optimization with the Alternating Direction Method of Multipliers

Simple item page
dc.contributor.authorShi, W.
dc.contributor.authorLing, Q.
dc.contributor.authorYuan, K.
dc.contributor.authorWu, G.
dc.contributor.authorYin, W.
dc.date.accessioned2018-06-19T17:48:49Z
dc.date.available2018-06-19T17:48:49Z
dc.date.issued2013-04
dc.date.noteApril 2013
dc.description.abstractIn a decentralized consensus optimization problem, a network of agents minimizes the summation of their local objective functions on a common set of decision variables, allowing only information exchange among neighbors. The alternating direction method of multipliers (ADMM) has been shown to be a powerful tool for solving the problem with empirically fast convergence. This paper establishes the linear convergence rate of the ADMM in decentralized consensus optimization. The theoretical convergence rate is a function of the network topology, properties of the local objective functions, and the algorithm parameter. This result not only gives a performance guarantee for the ADMM but also provides a guideline to accelerate its convergence for decentralized consensus optimization problems.
dc.format.extent5 pp
dc.identifier.citationShi, W., Ling, Q., Yuan, K., et al.. "Linearly Convergent Decentralized Consensus Optimization with the Alternating Direction Method of Multipliers." (2013) <a href="https://hdl.handle.net/1911/102219">https://hdl.handle.net/1911/102219</a>.
dc.identifier.digitalTR13-07
dc.identifier.urihttps://hdl.handle.net/1911/102219
dc.language.isoeng
dc.titleLinearly Convergent Decentralized Consensus Optimization with the Alternating Direction Method of Multipliers
dc.typeTechnical report
dc.type.dcmiText
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
TR13-07.PDF
Size:
179.41 KB
Format:
Adobe Portable Document Format
Download
Collections
CAAM Technical Reports