Adaptation and learning over networks under subspace constraints-part II: performance analysis
OA Location
Author(s)
Nassif, Roula
Vlaski, Stefan
Sayed, Ali H
Type
Journal Article
Abstract
Part I of this paper considered optimization problems over networks where agents have individual objectives to meet, or individual parameter vectors to estimate, subject to subspace constraints that require the objectives across the network to lie in low-dimensional subspaces. Starting from the centralized projected gradient descent, an iterative and distributed solution was proposed that responds to streaming data and employs stochastic approximations in place of actual gradient vectors, which are generally unavailable. We examined the second-order stability of the learning algorithm and we showed that, for small step-sizes μ, the proposed strategy leads to small estimation errors on the order of μ. This Part II examines steady-state performance. The results reveal explicitly the influence of the gradient noise, data characteristics, and subspace constraints, on the network performance. The results also show that in the small step-size regime, the iterates generated by the distributed algorithm achieve the centralized steady-state performance.
Date Issued
2020
Date Acceptance
2020-04-05
Citation
IEEE Transactions on Signal Processing, 2020, 68, pp.2948-2962
ISSN
1053-587X
Publisher
Institute of Electrical and Electronics Engineers
Start Page
2948
End Page
2962
Journal / Book Title
IEEE Transactions on Signal Processing
Volume
68
Copyright Statement
© 2020 IEEE
Identifier
https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000538022500002&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=a2bf6146997ec60c407a63945d4e92bb
Subjects
CONSENSUS
Covariance matrices
DIFFUSION LMS
Distributed optimization
Engineering
Engineering, Electrical & Electronic
gradient noise
Network topology
Optimization
PROJECTION ALGORITHMS
Science & Technology
SENSOR NETWORKS
Signal processing algorithms
Stability analysis
Steady-state
steady-state performance
Subspace constraints
subspace projection
Technology
Publication Status
Published
Date Publish Online
2020-04-15