A sparsity promoting adaptive algorithm for distributed learning

Symeon Chouvardas; Konstantinos Slavakis; Yannis Kopsinis; Sergios Theodoridis

doi:10.1109/TSP.2012.2204987

A sparsity promoting adaptive algorithm for distributed learning

Symeon Chouvardas, Konstantinos Slavakis, Yannis Kopsinis, Sergios Theodoridis

Digital Technology Center

Research output: Contribution to journal › Article › peer-review

74 Scopus citations

Abstract

In this paper, a sparsity promoting adaptive algorithm for distributed learning in diffusion networks is developed. The algorithm follows the set-theoretic estimation rationale. At each time instance and at each node of the network, a closed convex set, known as property set, is constructed based on the received measurements; this defines the region in which the solution is searched for. In this paper, the property sets take the form of hyperslabs. The goal is to find a point that belongs to the intersection of these hyperslabs. To this end, sparsity encouraging variable metric projections onto the hyperslabs have been adopted. In addition, sparsity is also imposed by employing variable metric projections onto weighted balls. A combine adapt cooperation strategy is adopted. Under some mild assumptions, the scheme enjoys monotonicity, asymptotic optimality and strong convergence to a point that lies in the consensus subspace. Finally, numerical examples verify the validity of the proposed scheme compared to other algorithms, which have been developed in the context of sparse adaptive learning.

Original language	English (US)
Article number	6218786
Pages (from-to)	5412-5424
Number of pages	13
Journal	IEEE Transactions on Signal Processing
Volume	60
Issue number	10
DOIs	https://doi.org/10.1109/TSP.2012.2204987
State	Published - 2012

Bibliographical note

Funding Information:
Manuscript received November 08, 2011; revised March 17, 2012 and May 28, 2012; accepted May 30, 2012. Date of publication June 15, 2012; date of current version September 11, 2012. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Huaiyu Dai. This work was supported by the European Union (European Social Fund—ESF) and by the Greek national funds through the Operational Program “Education and Lifelong Learning” of the National Strategic Reference Framework (NSRF)—Research Funding Program: Heracleitus II. Investing in knowledge society through the European Social Fund.

Keywords

Adaptive distributed learning
Diffusion networks
Projections
Sparsity

Access

10.1109/TSP.2012.2204987

OpenUrl availability

Full text

Cite this

@article{97ce41ea4f6145a4b8bc16edabfa85a2,

title = "A sparsity promoting adaptive algorithm for distributed learning",

abstract = "In this paper, a sparsity promoting adaptive algorithm for distributed learning in diffusion networks is developed. The algorithm follows the set-theoretic estimation rationale. At each time instance and at each node of the network, a closed convex set, known as property set, is constructed based on the received measurements; this defines the region in which the solution is searched for. In this paper, the property sets take the form of hyperslabs. The goal is to find a point that belongs to the intersection of these hyperslabs. To this end, sparsity encouraging variable metric projections onto the hyperslabs have been adopted. In addition, sparsity is also imposed by employing variable metric projections onto weighted balls. A combine adapt cooperation strategy is adopted. Under some mild assumptions, the scheme enjoys monotonicity, asymptotic optimality and strong convergence to a point that lies in the consensus subspace. Finally, numerical examples verify the validity of the proposed scheme compared to other algorithms, which have been developed in the context of sparse adaptive learning.",

keywords = "Adaptive distributed learning, Diffusion networks, Projections, Sparsity",

author = "Symeon Chouvardas and Konstantinos Slavakis and Yannis Kopsinis and Sergios Theodoridis",

note = "Funding Information: Manuscript received November 08, 2011; revised March 17, 2012 and May 28, 2012; accepted May 30, 2012. Date of publication June 15, 2012; date of current version September 11, 2012. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Huaiyu Dai. This work was supported by the European Union (European Social Fund—ESF) and by the Greek national funds through the Operational Program “Education and Lifelong Learning” of the National Strategic Reference Framework (NSRF)—Research Funding Program: Heracleitus II. Investing in knowledge society through the European Social Fund.",

year = "2012",

doi = "10.1109/TSP.2012.2204987",

language = "English (US)",

volume = "60",

pages = "5412--5424",

journal = "IEEE Transactions on Signal Processing",

issn = "1053-587X",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "10",

}

TY - JOUR

T1 - A sparsity promoting adaptive algorithm for distributed learning

AU - Chouvardas, Symeon

AU - Slavakis, Konstantinos

AU - Kopsinis, Yannis

AU - Theodoridis, Sergios

N1 - Funding Information: Manuscript received November 08, 2011; revised March 17, 2012 and May 28, 2012; accepted May 30, 2012. Date of publication June 15, 2012; date of current version September 11, 2012. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Huaiyu Dai. This work was supported by the European Union (European Social Fund—ESF) and by the Greek national funds through the Operational Program “Education and Lifelong Learning” of the National Strategic Reference Framework (NSRF)—Research Funding Program: Heracleitus II. Investing in knowledge society through the European Social Fund.

PY - 2012

Y1 - 2012

N2 - In this paper, a sparsity promoting adaptive algorithm for distributed learning in diffusion networks is developed. The algorithm follows the set-theoretic estimation rationale. At each time instance and at each node of the network, a closed convex set, known as property set, is constructed based on the received measurements; this defines the region in which the solution is searched for. In this paper, the property sets take the form of hyperslabs. The goal is to find a point that belongs to the intersection of these hyperslabs. To this end, sparsity encouraging variable metric projections onto the hyperslabs have been adopted. In addition, sparsity is also imposed by employing variable metric projections onto weighted balls. A combine adapt cooperation strategy is adopted. Under some mild assumptions, the scheme enjoys monotonicity, asymptotic optimality and strong convergence to a point that lies in the consensus subspace. Finally, numerical examples verify the validity of the proposed scheme compared to other algorithms, which have been developed in the context of sparse adaptive learning.

AB - In this paper, a sparsity promoting adaptive algorithm for distributed learning in diffusion networks is developed. The algorithm follows the set-theoretic estimation rationale. At each time instance and at each node of the network, a closed convex set, known as property set, is constructed based on the received measurements; this defines the region in which the solution is searched for. In this paper, the property sets take the form of hyperslabs. The goal is to find a point that belongs to the intersection of these hyperslabs. To this end, sparsity encouraging variable metric projections onto the hyperslabs have been adopted. In addition, sparsity is also imposed by employing variable metric projections onto weighted balls. A combine adapt cooperation strategy is adopted. Under some mild assumptions, the scheme enjoys monotonicity, asymptotic optimality and strong convergence to a point that lies in the consensus subspace. Finally, numerical examples verify the validity of the proposed scheme compared to other algorithms, which have been developed in the context of sparse adaptive learning.

KW - Adaptive distributed learning

KW - Diffusion networks

KW - Projections

KW - Sparsity

UR - http://www.scopus.com/inward/record.url?scp=84866481291&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866481291&partnerID=8YFLogxK

U2 - 10.1109/TSP.2012.2204987

DO - 10.1109/TSP.2012.2204987

M3 - Article

AN - SCOPUS:84866481291

SN - 1053-587X

VL - 60

SP - 5412

EP - 5424

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

IS - 10

M1 - 6218786

ER -

A sparsity promoting adaptive algorithm for distributed learning

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this