Maximum a posteriori vs maximum probability recursive sparse estimation

Lachlan Blackhall; Michael Rotkowitz

doi:10.23919/ecc.2009.7074447

Maximum a posteriori vs maximum probability recursive sparse estimation

Lachlan Blackhall, Michael Rotkowitz

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Recursive sparse parameter estimates obtained using the author's recent maximum a posteriori (MAP) approach, where the sparse parameter estimates are determined as the a posteriori mode of a Gaussian sum filter, are compared with a new maximum probability (MP) methodology, where the sparse parameter estimates are determined as the component of a Gaussian sum filter with the highest a posteriori weighting. We show how the performance of the MP estimator approach to sparse parameter estimates, in both sparsity and mean square error senses, depends on the parameters that characterize each multivariate Gaussian in the Gaussian sum filter. Through this work we also provide additional performance analysis for the MP estimator and suggest possible areas of future work that will further improve its performance.

Original language	English
Title of host publication	2009 European Control Conference, ECC 2009
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	472-477
Number of pages	6
ISBN (Electronic)	9783952417393
DOIs	https://doi.org/10.23919/ecc.2009.7074447
Publication status	Published - 26 Mar 2014
Event	10th European Control Conference: ECC 2009 - Budapest, Hungary Duration: 23 Aug 2009 → 26 Aug 2009 Conference number: 10th http://ieeexplore.ieee.org/document/7074402/ http://ieeexplore.ieee.org/document/7074461/

Publication series

Name	2009 European Control Conference, ECC 2009

Conference

Conference	10th European Control Conference: ECC 2009
Abbreviated title	ECC
Country/Territory	Hungary
City	Budapest
Period	23/08/09 → 26/08/09
Internet address	http://ieeexplore.ieee.org/document/7074402/ http://ieeexplore.ieee.org/document/7074461/

Access to Document

10.23919/ecc.2009.7074447

Cite this

@inproceedings{2916bbdcc0eb458298302274c35bb700,

title = "Maximum a posteriori vs maximum probability recursive sparse estimation",

abstract = "Recursive sparse parameter estimates obtained using the author's recent maximum a posteriori (MAP) approach, where the sparse parameter estimates are determined as the a posteriori mode of a Gaussian sum filter, are compared with a new maximum probability (MP) methodology, where the sparse parameter estimates are determined as the component of a Gaussian sum filter with the highest a posteriori weighting. We show how the performance of the MP estimator approach to sparse parameter estimates, in both sparsity and mean square error senses, depends on the parameters that characterize each multivariate Gaussian in the Gaussian sum filter. Through this work we also provide additional performance analysis for the MP estimator and suggest possible areas of future work that will further improve its performance.",

author = "Lachlan Blackhall and Michael Rotkowitz",

note = "Publisher Copyright: {\textcopyright} 2009 EUCA.; 10th European Control Conference: ECC 2009, ECC ; Conference date: 23-08-2009 Through 26-08-2009",

year = "2014",

month = mar,

day = "26",

doi = "10.23919/ecc.2009.7074447",

language = "English",

series = "2009 European Control Conference, ECC 2009",

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

pages = "472--477",

booktitle = "2009 European Control Conference, ECC 2009",

address = "United States",

url = "http://ieeexplore.ieee.org/document/7074402/, http://ieeexplore.ieee.org/document/7074461/",

}

Blackhall, L & Rotkowitz, M 2014, Maximum a posteriori vs maximum probability recursive sparse estimation. in 2009 European Control Conference, ECC 2009., 7074447, 2009 European Control Conference, ECC 2009, Institute of Electrical and Electronics Engineers Inc., pp. 472-477, 10th European Control Conference: ECC 2009, Budapest, Hungary, 23/08/09. https://doi.org/10.23919/ecc.2009.7074447

Maximum a posteriori vs maximum probability recursive sparse estimation. / Blackhall, Lachlan; Rotkowitz, Michael.
2009 European Control Conference, ECC 2009. Institute of Electrical and Electronics Engineers Inc., 2014. p. 472-477 7074447 (2009 European Control Conference, ECC 2009).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Maximum a posteriori vs maximum probability recursive sparse estimation

AU - Blackhall, Lachlan

AU - Rotkowitz, Michael

N1 - Conference code: 10th

PY - 2014/3/26

Y1 - 2014/3/26

N2 - Recursive sparse parameter estimates obtained using the author's recent maximum a posteriori (MAP) approach, where the sparse parameter estimates are determined as the a posteriori mode of a Gaussian sum filter, are compared with a new maximum probability (MP) methodology, where the sparse parameter estimates are determined as the component of a Gaussian sum filter with the highest a posteriori weighting. We show how the performance of the MP estimator approach to sparse parameter estimates, in both sparsity and mean square error senses, depends on the parameters that characterize each multivariate Gaussian in the Gaussian sum filter. Through this work we also provide additional performance analysis for the MP estimator and suggest possible areas of future work that will further improve its performance.

AB - Recursive sparse parameter estimates obtained using the author's recent maximum a posteriori (MAP) approach, where the sparse parameter estimates are determined as the a posteriori mode of a Gaussian sum filter, are compared with a new maximum probability (MP) methodology, where the sparse parameter estimates are determined as the component of a Gaussian sum filter with the highest a posteriori weighting. We show how the performance of the MP estimator approach to sparse parameter estimates, in both sparsity and mean square error senses, depends on the parameters that characterize each multivariate Gaussian in the Gaussian sum filter. Through this work we also provide additional performance analysis for the MP estimator and suggest possible areas of future work that will further improve its performance.

UR - http://www.scopus.com/inward/record.url?scp=84955181608&partnerID=8YFLogxK

U2 - 10.23919/ecc.2009.7074447

DO - 10.23919/ecc.2009.7074447

M3 - Conference contribution

AN - SCOPUS:84955181608

T3 - 2009 European Control Conference, ECC 2009

SP - 472

EP - 477

BT - 2009 European Control Conference, ECC 2009

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 10th European Control Conference: ECC 2009

Y2 - 23 August 2009 through 26 August 2009

ER -

Maximum a posteriori vs maximum probability recursive sparse estimation

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this