Scalable graph signal recovery for big data over networks

Alex Jung; Peter Berger; Gabor Hannak; Gerald Matz

doi:10.1109/SPAWC.2016.7536869

Scalable graph signal recovery for big data over networks

Alex Jung
, Peter Berger
, Gabor Hannak
, Gerald Matz

Vienna University of Technology

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

9 Citations (Scopus)

Abstract

We formulate the recovery of a graph signal from noisy samples taken on a subset of graph nodes as a convex optimization problem that balances the empirical error for explaining the observed values and a complexity term quantifying the smoothness of the graph signal. To solve this optimization problem, we propose to combine the alternating direction method of multipliers with a novel denoising method that minimizes total variation. Our algorithm can be efficiently implemented in a distributed manner using message passing and thus is attractive for big data problems over networks.

Original language	English
Title of host publication	SPAWC 2016 - 17th IEEE International Workshop on Signal Processing Advances in Wireless Communications
Publisher	IEEE
Number of pages	6
Volume	2016-August
ISBN (Electronic)	9781509017492
DOIs	https://doi.org/10.1109/SPAWC.2016.7536869
Publication status	Published - 9 Aug 2016
MoE publication type	A4 Conference publication
Event	IEEE International Workshop on Signal Processing Advances in Wireless Communications - Edinburgh, United Kingdom Duration: 3 Jul 2016 → 6 Jul 2016 Conference number: 17

Workshop

Workshop	IEEE International Workshop on Signal Processing Advances in Wireless Communications
Abbreviated title	SPAWC
Country/Territory	United Kingdom
City	Edinburgh
Period	03/07/2016 → 06/07/2016

Keywords

ADMM
big data
graph signal recovery
message passing
total variation

Access to Document

10.1109/SPAWC.2016.7536869

Cite this

@inproceedings{becbfa726fa1443aa380d469b32aacf1,

title = "Scalable graph signal recovery for big data over networks",

abstract = "We formulate the recovery of a graph signal from noisy samples taken on a subset of graph nodes as a convex optimization problem that balances the empirical error for explaining the observed values and a complexity term quantifying the smoothness of the graph signal. To solve this optimization problem, we propose to combine the alternating direction method of multipliers with a novel denoising method that minimizes total variation. Our algorithm can be efficiently implemented in a distributed manner using message passing and thus is attractive for big data problems over networks.",

keywords = "ADMM, big data, graph signal recovery, message passing, total variation",

author = "Alex Jung and Peter Berger and Gabor Hannak and Gerald Matz",

year = "2016",

month = aug,

day = "9",

doi = "10.1109/SPAWC.2016.7536869",

language = "English",

volume = "2016-August",

booktitle = "SPAWC 2016 - 17th IEEE International Workshop on Signal Processing Advances in Wireless Communications",

publisher = "IEEE",

address = "United States",

note = "IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC ; Conference date: 03-07-2016 Through 06-07-2016",

}

Jung, A, Berger, P, Hannak, G & Matz, G 2016, Scalable graph signal recovery for big data over networks. in SPAWC 2016 - 17th IEEE International Workshop on Signal Processing Advances in Wireless Communications. vol. 2016-August, 7536869, IEEE, IEEE International Workshop on Signal Processing Advances in Wireless Communications, Edinburgh, United Kingdom, 03/07/2016. https://doi.org/10.1109/SPAWC.2016.7536869

Scalable graph signal recovery for big data over networks. / Jung, Alex; Berger, Peter; Hannak, Gabor et al.
SPAWC 2016 - 17th IEEE International Workshop on Signal Processing Advances in Wireless Communications. Vol. 2016-August IEEE, 2016. 7536869.

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

TY - GEN

T1 - Scalable graph signal recovery for big data over networks

AU - Jung, Alex

AU - Berger, Peter

AU - Hannak, Gabor

AU - Matz, Gerald

N1 - Conference code: 17

PY - 2016/8/9

Y1 - 2016/8/9

N2 - We formulate the recovery of a graph signal from noisy samples taken on a subset of graph nodes as a convex optimization problem that balances the empirical error for explaining the observed values and a complexity term quantifying the smoothness of the graph signal. To solve this optimization problem, we propose to combine the alternating direction method of multipliers with a novel denoising method that minimizes total variation. Our algorithm can be efficiently implemented in a distributed manner using message passing and thus is attractive for big data problems over networks.

AB - We formulate the recovery of a graph signal from noisy samples taken on a subset of graph nodes as a convex optimization problem that balances the empirical error for explaining the observed values and a complexity term quantifying the smoothness of the graph signal. To solve this optimization problem, we propose to combine the alternating direction method of multipliers with a novel denoising method that minimizes total variation. Our algorithm can be efficiently implemented in a distributed manner using message passing and thus is attractive for big data problems over networks.

KW - ADMM

KW - big data

KW - graph signal recovery

KW - message passing

KW - total variation

UR - https://www.scopus.com/pages/publications/84984619771

U2 - 10.1109/SPAWC.2016.7536869

DO - 10.1109/SPAWC.2016.7536869

M3 - Conference article in proceedings

AN - SCOPUS:84984619771

VL - 2016-August

BT - SPAWC 2016 - 17th IEEE International Workshop on Signal Processing Advances in Wireless Communications

PB - IEEE

T2 - IEEE International Workshop on Signal Processing Advances in Wireless Communications

Y2 - 3 July 2016 through 6 July 2016

ER -

Scalable graph signal recovery for big data over networks

Abstract

Workshop

Keywords

Access to Document

Other files and links

Fingerprint

Cite this