A Network Compatibility Condition for Compressed Sensing over Complex Networks

Nguyen Tran; Henrik Ambos; Alexander Jung

doi:10.1109/SSP.2018.8450811

A Network Compatibility Condition for Compressed Sensing over Complex Networks

Nguyen Tran, Henrik Ambos, Alexander Jung

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

1 Citation (Scopus)

201 Downloads (Pure)

Abstract

This paper continues our recently initiated line of work on analyzing the network Lasso (nLasso, which has been proposed as an efficient learning algorithm for massive networkstructured data sets (big data over networks). The nLasso extends the well-known Lasso estimator to network-structured datasets. In this paper we consider the nLasso using squared error loss and provide sufficient conditions on the network structure and available label information such that nLasso accurately recovers a clustered (piece-wise constant) graph signal (representing label information) from the information pro-vided by the labels of a few data points.

Original language	English
Title of host publication	2018 IEEE Statistical Signal Processing Workshop, SSP 2018
Publisher	IEEE
Pages	50-54
Number of pages	5
ISBN (Print)	9781538615706
DOIs	https://doi.org/10.1109/SSP.2018.8450811
Publication status	Published - 29 Aug 2018
MoE publication type	A4 Conference publication
Event	IEEE Statistical Signal Processing Workshop - Freiburg im Breisgau, Germany Duration: 10 Jun 2018 → 13 Jun 2018 Conference number: 20

Workshop

Workshop	IEEE Statistical Signal Processing Workshop
Abbreviated title	SSP
Country/Territory	Germany
City	Freiburg im Breisgau
Period	10/06/2018 → 13/06/2018

Keywords

big data over networks
complex networks
compressed sensing
network compatibility condition
network Lasso

Access to Document

10.1109/SSP.2018.8450811

SCI_Tran_A_Network_Compatibility_nguyen_sspAccepted author manuscript, 479 KB

Cite this

@inproceedings{30d4ebdb883d431cb944ebad9b08d0ea,

title = "A Network Compatibility Condition for Compressed Sensing over Complex Networks",

abstract = "This paper continues our recently initiated line of work on analyzing the network Lasso (nLasso, which has been proposed as an efficient learning algorithm for massive networkstructured data sets (big data over networks). The nLasso extends the well-known Lasso estimator to network-structured datasets. In this paper we consider the nLasso using squared error loss and provide sufficient conditions on the network structure and available label information such that nLasso accurately recovers a clustered (piece-wise constant) graph signal (representing label information) from the information pro-vided by the labels of a few data points.",

keywords = "big data over networks, complex networks, compressed sensing, network compatibility condition, network Lasso",

author = "Nguyen Tran and Henrik Ambos and Alexander Jung",

year = "2018",

month = aug,

day = "29",

doi = "10.1109/SSP.2018.8450811",

language = "English",

isbn = "9781538615706",

pages = "50--54",

booktitle = "2018 IEEE Statistical Signal Processing Workshop, SSP 2018",

publisher = "IEEE",

address = "United States",

note = "IEEE Statistical Signal Processing Workshop, SSP ; Conference date: 10-06-2018 Through 13-06-2018",

}

TY - GEN

T1 - A Network Compatibility Condition for Compressed Sensing over Complex Networks

AU - Tran, Nguyen

AU - Ambos, Henrik

AU - Jung, Alexander

N1 - Conference code: 20

PY - 2018/8/29

Y1 - 2018/8/29

N2 - This paper continues our recently initiated line of work on analyzing the network Lasso (nLasso, which has been proposed as an efficient learning algorithm for massive networkstructured data sets (big data over networks). The nLasso extends the well-known Lasso estimator to network-structured datasets. In this paper we consider the nLasso using squared error loss and provide sufficient conditions on the network structure and available label information such that nLasso accurately recovers a clustered (piece-wise constant) graph signal (representing label information) from the information pro-vided by the labels of a few data points.

AB - This paper continues our recently initiated line of work on analyzing the network Lasso (nLasso, which has been proposed as an efficient learning algorithm for massive networkstructured data sets (big data over networks). The nLasso extends the well-known Lasso estimator to network-structured datasets. In this paper we consider the nLasso using squared error loss and provide sufficient conditions on the network structure and available label information such that nLasso accurately recovers a clustered (piece-wise constant) graph signal (representing label information) from the information pro-vided by the labels of a few data points.

KW - big data over networks

KW - complex networks

KW - compressed sensing

KW - network compatibility condition

KW - network Lasso

UR - http://www.scopus.com/inward/record.url?scp=85053856553&partnerID=8YFLogxK

U2 - 10.1109/SSP.2018.8450811

DO - 10.1109/SSP.2018.8450811

M3 - Conference article in proceedings

AN - SCOPUS:85053856553

SN - 9781538615706

SP - 50

EP - 54

BT - 2018 IEEE Statistical Signal Processing Workshop, SSP 2018

PB - IEEE

T2 - IEEE Statistical Signal Processing Workshop

Y2 - 10 June 2018 through 13 June 2018

ER -

A Network Compatibility Condition for Compressed Sensing over Complex Networks

Abstract

Workshop

Keywords

Access to Document

Other files and links

Fingerprint

Cite this