Regression with Sparse Approximations of Data

Pardis Noorzad, Bob L. Sturm

Publikation: Bidrag til tidsskriftKonferenceartikel i tidsskriftForskningpeer review

5 Citationer (Scopus)
330 Downloads (Pure)

Resumé

We propose sparse approximation weighted regression (SPARROW), a method for local estimation of the regression function that uses sparse approximation with a dictionary of measurements. SPARROW estimates the regression function at a point with a linear combination of a few regressands selected by a sparse approximation of the point in terms of the regressors. We show SPARROW can be considered a variant of \(k\)-nearest neighbors regression (\(k\)-NNR), and more generally, local polynomial kernel regression. Unlike \(k\)-NNR, however, SPARROW can adapt the number of regressors to use based on the sparse approximation process. Our experimental results show the locally constant form of SPARROW performs competitively.
OriginalsprogEngelsk
TidsskriftProceedings of the European Signal Processing Conference
Vol/bind2012
Sider (fra-til)674-678
Antal sider5
ISSN2076-1465
StatusUdgivet - 2012
BegivenhedEUSIPCO2012 - Bucharest, Rumænien
Varighed: 27 aug. 2012 → …

Konference

KonferenceEUSIPCO2012
LandRumænien
ByBucharest
Periode27/08/2012 → …

Fingerprint

Glossaries
Polynomials

Citer dette

Noorzad, Pardis ; Sturm, Bob L. / Regression with Sparse Approximations of Data. I: Proceedings of the European Signal Processing Conference. 2012 ; Bind 2012. s. 674-678 .
@inproceedings{87d9a6cf824d4006b7be23fed96b6f58,
title = "Regression with Sparse Approximations of Data",
abstract = "We propose sparse approximation weighted regression (SPARROW), a method for local estimation of the regression function that uses sparse approximation with a dictionary of measurements. SPARROW estimates the regression function at a point with a linear combination of a few regressands selected by a sparse approximation of the point in terms of the regressors. We show SPARROW can be considered a variant of \(k\)-nearest neighbors regression (\(k\)-NNR), and more generally, local polynomial kernel regression. Unlike \(k\)-NNR, however, SPARROW can adapt the number of regressors to use based on the sparse approximation process. Our experimental results show the locally constant form of SPARROW performs competitively.",
author = "Pardis Noorzad and Sturm, {Bob L.}",
year = "2012",
language = "English",
volume = "2012",
pages = "674--678",
journal = "Proceedings of the European Signal Processing Conference",
issn = "2076-1465",
publisher = "European Association for Signal Processing (EURASIP)",

}

Regression with Sparse Approximations of Data. / Noorzad, Pardis; Sturm, Bob L.

I: Proceedings of the European Signal Processing Conference, Bind 2012, 2012, s. 674-678 .

Publikation: Bidrag til tidsskriftKonferenceartikel i tidsskriftForskningpeer review

TY - GEN

T1 - Regression with Sparse Approximations of Data

AU - Noorzad, Pardis

AU - Sturm, Bob L.

PY - 2012

Y1 - 2012

N2 - We propose sparse approximation weighted regression (SPARROW), a method for local estimation of the regression function that uses sparse approximation with a dictionary of measurements. SPARROW estimates the regression function at a point with a linear combination of a few regressands selected by a sparse approximation of the point in terms of the regressors. We show SPARROW can be considered a variant of \(k\)-nearest neighbors regression (\(k\)-NNR), and more generally, local polynomial kernel regression. Unlike \(k\)-NNR, however, SPARROW can adapt the number of regressors to use based on the sparse approximation process. Our experimental results show the locally constant form of SPARROW performs competitively.

AB - We propose sparse approximation weighted regression (SPARROW), a method for local estimation of the regression function that uses sparse approximation with a dictionary of measurements. SPARROW estimates the regression function at a point with a linear combination of a few regressands selected by a sparse approximation of the point in terms of the regressors. We show SPARROW can be considered a variant of \(k\)-nearest neighbors regression (\(k\)-NNR), and more generally, local polynomial kernel regression. Unlike \(k\)-NNR, however, SPARROW can adapt the number of regressors to use based on the sparse approximation process. Our experimental results show the locally constant form of SPARROW performs competitively.

UR - http://www.scopus.com/inward/record.url?scp=84869790257&partnerID=8YFLogxK

M3 - Conference article in Journal

VL - 2012

SP - 674

EP - 678

JO - Proceedings of the European Signal Processing Conference

JF - Proceedings of the European Signal Processing Conference

SN - 2076-1465

ER -