Distributed Splitting-Over-Features Sparse Bayesian Learning with Alternating Direction Method of Multipliers

Christoph Manss; Dmitriy Shutin; Geert Leus

doi:10.1109/ICASSP.2018.8462229

Distributed Splitting-Over-Features Sparse Bayesian Learning with Alternating Direction Method of Multipliers

Christoph Manss, Dmitriy Shutin, Geert Leus

Signal Processing Systems

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

6 Citations (Scopus)

56 Downloads (Pure)

Abstract

In processing spatially distributed data, multi-agent robotic platforms equipped with sensors and computing capabilities are gaining interest for applications in inhospitable environments. In this work an algorithm for a distributed realization of sparse bayesian learning (SBL) is discussed for learning a static spatial process with the splitting-over-features approach over a network of interconnected agents. The observed process is modeled as a superposition of weighted kernel functions, or features as we call it, centered at the agent's measurement locations. SBL is then used to determine which feature is relevant for representing the spatial process. Using upper bounding convex functions, the SBL parameter estimation is formulated as ℓ1-norm constrained optimization, which is solved distributively using alternating direction method of multipliers (ADMM) and averaged consensus. The performance of the method is demonstrated by processing real magnetic field data collected in a laboratory.

Original language	English
Title of host publication	2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings
Place of Publication	Piscataway, NJ
Publisher	IEEE
Pages	3654-3658
Number of pages	5
ISBN (Electronic)	978-1-5386-4658-8
ISBN (Print)	978-1-5386-4659-5
DOIs	https://doi.org/10.1109/ICASSP.2018.8462229
Publication status	Published - 2018
Event	2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018: Signal Processing and Artificial Intelligence: Changing the World - Calgary Telus Convention Center (CTCC), Calgary, Canada Duration: 15 Apr 2018 → 20 Apr 2018 https://2018.ieeeicassp.org

Conference

Conference	2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018
Country/Territory	Canada
City	Calgary
Period	15/04/18 → 20/04/18
Internet address	https://2018.ieeeicassp.org

Bibliographical note

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

ADMM
Learning over networks
Multi-agent systems
Sparse Bayesian learning

Access to Document

10.1109/ICASSP.2018.8462229

Distributed_Splitting-Over-Features_Sparse_Bayesian_Learning_with_Alternating_Direction_Method_of_MultipliersFinal published version, 671 KB

Cite this

@inproceedings{28f14cce34af4b6a8e88823c3763d046,

title = "Distributed Splitting-Over-Features Sparse Bayesian Learning with Alternating Direction Method of Multipliers",

abstract = "In processing spatially distributed data, multi-agent robotic platforms equipped with sensors and computing capabilities are gaining interest for applications in inhospitable environments. In this work an algorithm for a distributed realization of sparse bayesian learning (SBL) is discussed for learning a static spatial process with the splitting-over-features approach over a network of interconnected agents. The observed process is modeled as a superposition of weighted kernel functions, or features as we call it, centered at the agent's measurement locations. SBL is then used to determine which feature is relevant for representing the spatial process. Using upper bounding convex functions, the SBL parameter estimation is formulated as ℓ1-norm constrained optimization, which is solved distributively using alternating direction method of multipliers (ADMM) and averaged consensus. The performance of the method is demonstrated by processing real magnetic field data collected in a laboratory.",

keywords = "ADMM, Learning over networks, Multi-agent systems, Sparse Bayesian learning",

author = "Christoph Manss and Dmitriy Shutin and Geert Leus",

note = "Green Open Access added to TU Delft Institutional Repository {\textquoteleft}You share, we take care!{\textquoteright} – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.; 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 : Signal Processing and Artificial Intelligence: Changing the World ; Conference date: 15-04-2018 Through 20-04-2018",

year = "2018",

doi = "10.1109/ICASSP.2018.8462229",

language = "English",

isbn = "978-1-5386-4659-5",

pages = "3654--3658",

booktitle = "2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings",

publisher = "IEEE",

address = "United States",

url = "https://2018.ieeeicassp.org",

}

Manss, C, Shutin, D & Leus, G 2018, Distributed Splitting-Over-Features Sparse Bayesian Learning with Alternating Direction Method of Multipliers. in 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings., 8462229, IEEE, Piscataway, NJ, pp. 3654-3658, 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018, Calgary, Canada, 15/04/18. https://doi.org/10.1109/ICASSP.2018.8462229

Distributed Splitting-Over-Features Sparse Bayesian Learning with Alternating Direction Method of Multipliers. / Manss, Christoph; Shutin, Dmitriy; Leus, Geert.
2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings. Piscataway, NJ: IEEE, 2018. p. 3654-3658 8462229.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Distributed Splitting-Over-Features Sparse Bayesian Learning with Alternating Direction Method of Multipliers

AU - Manss, Christoph

AU - Shutin, Dmitriy

AU - Leus, Geert

N1 - Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2018

Y1 - 2018

N2 - In processing spatially distributed data, multi-agent robotic platforms equipped with sensors and computing capabilities are gaining interest for applications in inhospitable environments. In this work an algorithm for a distributed realization of sparse bayesian learning (SBL) is discussed for learning a static spatial process with the splitting-over-features approach over a network of interconnected agents. The observed process is modeled as a superposition of weighted kernel functions, or features as we call it, centered at the agent's measurement locations. SBL is then used to determine which feature is relevant for representing the spatial process. Using upper bounding convex functions, the SBL parameter estimation is formulated as ℓ1-norm constrained optimization, which is solved distributively using alternating direction method of multipliers (ADMM) and averaged consensus. The performance of the method is demonstrated by processing real magnetic field data collected in a laboratory.

AB - In processing spatially distributed data, multi-agent robotic platforms equipped with sensors and computing capabilities are gaining interest for applications in inhospitable environments. In this work an algorithm for a distributed realization of sparse bayesian learning (SBL) is discussed for learning a static spatial process with the splitting-over-features approach over a network of interconnected agents. The observed process is modeled as a superposition of weighted kernel functions, or features as we call it, centered at the agent's measurement locations. SBL is then used to determine which feature is relevant for representing the spatial process. Using upper bounding convex functions, the SBL parameter estimation is formulated as ℓ1-norm constrained optimization, which is solved distributively using alternating direction method of multipliers (ADMM) and averaged consensus. The performance of the method is demonstrated by processing real magnetic field data collected in a laboratory.

KW - ADMM

KW - Learning over networks

KW - Multi-agent systems

KW - Sparse Bayesian learning

UR - http://www.scopus.com/inward/record.url?scp=85054202092&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2018.8462229

DO - 10.1109/ICASSP.2018.8462229

M3 - Conference contribution

AN - SCOPUS:85054202092

SN - 978-1-5386-4659-5

SP - 3654

EP - 3658

BT - 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings

PB - IEEE

CY - Piscataway, NJ

T2 - 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018

Y2 - 15 April 2018 through 20 April 2018

ER -

Distributed Splitting-Over-Features Sparse Bayesian Learning with Alternating Direction Method of Multipliers

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this