Regression with Sparse Approximations of Data

Publikation: Bidrag til tidsskriftKonferenceartikel i tidsskriftForskningpeer review

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.
Luk

Detaljer

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
Volume/Bind2012
Sider (fra-til)674-678
Antal sider5
ISSN2076-1465
StatusUdgivet - 2012
PublikationsartForskning
Peer reviewJa
BegivenhedEUSIPCO2012 - Bucharest, Rumænien
Varighed: 27 aug. 2012 → …

Konference

KonferenceEUSIPCO2012
LandRumænien
ByBucharest
Periode27/08/2012 → …

Download-statistik

Ingen data tilgængelig
ID: 71866590