Joint optimization of linear and nonlinear models for sequential regression

Simple item page
buir.contributor.authorFazla, Arda
buir.contributor.authorAydin, Mustafa E.
buir.contributor.authorKozat, Süleyman S.
buir.contributor.orcidFazla, Arda|0000-0001-6763-7266
buir.contributor.orcidKozat, Süleyman S.|0000-0002-6488-3848
dc.citation.epage103802-11en_US
dc.citation.spage103802-1en_US
dc.citation.volumeNumber132en_US
dc.contributor.authorFazla, Arda
dc.contributor.authorAydin, Mustafa E.
dc.contributor.authorKozat, Süleyman S.
dc.date.accessioned2023-02-16T12:29:02Z
dc.date.available2023-02-16T12:29:02Z
dc.date.issued2022-12
dc.departmentDepartment of Electrical and Electronics Engineeringen_US
dc.description.abstractWe investigate nonlinear regression and introduce a novel approach based on the joint optimization of linear and nonlinear models. In order to capture both the nonlinear and linear characteristics in sequential data, we model the underlying data as a combination of linear and nonlinear models, where we optimize the models jointly to minimize the final regression error. As the nonlinear model, we employ a differentiable version of the boosted decision trees. As the linear model, we use the well-known SARIMAX model. Our approach is generic so that any differentiable nonlinear or linear model can be readily employed provided that they are differentiable. By this joint optimization, we alleviate the well-known underfitting and overfitting problems in modeling sequential data. Through our experiments on synthetic and real-life data, we demonstrate significant improvements over individual components as well as the combination/mixture methods in the literature.en_US
dc.description.provenanceSubmitted by Bilge Kat (bilgekat@bilkent.edu.tr) on 2023-02-16T12:29:02Z No. of bitstreams: 1 Joint_optimization_of_linear_and_nonlinear_models_for_sequential_regression.pdf: 1596378 bytes, checksum: ceb6276a925476f39e5e00abc72a28cb (MD5)en
dc.description.provenanceMade available in DSpace on 2023-02-16T12:29:02Z (GMT). No. of bitstreams: 1 Joint_optimization_of_linear_and_nonlinear_models_for_sequential_regression.pdf: 1596378 bytes, checksum: ceb6276a925476f39e5e00abc72a28cb (MD5) Previous issue date: 2022-12en
dc.identifier.doi10.1016/j.dsp.2022.103802en_US
dc.identifier.eissn1095-4333
dc.identifier.issn1051-2004
dc.identifier.urihttp://hdl.handle.net/11693/111460
dc.language.isoEnglishen_US
dc.publisherAcademic Pressen_US
dc.relation.isversionofhttps://dx.doi.org/10.1016/j.dsp.2022.103802en_US
dc.source.titleDigital Signal Processingen_US
dc.subjectOnline learningen_US
dc.subjectRegressionen_US
dc.subjectSeasonal auto-regressive integrated moving average with eXogenous factors (SARIMAX)en_US
dc.subjectSoft gradient boosting decision tree (Soft GBDT)en_US
dc.subjectStochastic gradient descent (SGD)en_US
dc.titleJoint optimization of linear and nonlinear models for sequential regressionen_US
dc.typeArticleen_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Joint_optimization_of_linear_and_nonlinear_models_for_sequential_regression.pdf
Size:
1.52 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.69 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Scholarly Publications - Electrical and Electronics Engineering