SVM regression through variational methods and its sequential implementation

Junbin Gao; S.R. Gunn; C.J. Harris

doi:10.1016/S0925-2312(03)00365-5

SVM regression through variational methods and its sequential implementation

Junbin Gao
, S.R. Gunn
, C.J. Harris

Research output: Contribution to journal › Article › peer-review

30 Citations (Scopus)

Abstract

We consider an SVM regression model based on kernel methods with a Gaussian prior distribution over the network parameters. We show that the variational techniques can be utilised to obtain a closed form a posteriori distribution over the parameters given the data hence yielding an a posteriori predictive model.

Original language	English
Pages (from-to)	151-167
Number of pages	17
Journal	Neurocomputing
Volume	55
Issue number	1-2
DOIs	https://doi.org/10.1016/S0925-2312(03)00365-5
Publication status	Published - 2003

Access to Document

10.1016/S0925-2312(03)00365-5

Cite this

@article{925914119dc44705862dc60f12e7f7a7,

title = "SVM regression through variational methods and its sequential implementation",

abstract = "We consider an SVM regression model based on kernel methods with a Gaussian prior distribution over the network parameters. We show that the variational techniques can be utilised to obtain a closed form a posteriori distribution over the parameters given the data hence yielding an a posteriori predictive model.",

keywords = "Bayesian inference, Bayesian variational algorithm, EM algorithm, Regression, Support vector machine",

author = "Junbin Gao and S.R. Gunn and C.J. Harris",

note = "Imported on 12 Apr 2017 - DigiTool details were: Journal title (773t) = Neurocomputing. ISSNs: 0925-2312; ",

year = "2003",

doi = "10.1016/S0925-2312(03)00365-5",

language = "English",

volume = "55",

pages = "151--167",

journal = "Neurocomputing",

issn = "0925-2312",

publisher = "Elsevier",

number = "1-2",

}

TY - JOUR

T1 - SVM regression through variational methods and its sequential implementation

AU - Gao, Junbin

AU - Gunn, S.R.

AU - Harris, C.J.

N1 - Imported on 12 Apr 2017 - DigiTool details were: Journal title (773t) = Neurocomputing. ISSNs: 0925-2312;

PY - 2003

Y1 - 2003

N2 - We consider an SVM regression model based on kernel methods with a Gaussian prior distribution over the network parameters. We show that the variational techniques can be utilised to obtain a closed form a posteriori distribution over the parameters given the data hence yielding an a posteriori predictive model.

AB - We consider an SVM regression model based on kernel methods with a Gaussian prior distribution over the network parameters. We show that the variational techniques can be utilised to obtain a closed form a posteriori distribution over the parameters given the data hence yielding an a posteriori predictive model.

KW - Bayesian inference

KW - Bayesian variational algorithm

KW - EM algorithm

KW - Regression

KW - Support vector machine

U2 - 10.1016/S0925-2312(03)00365-5

DO - 10.1016/S0925-2312(03)00365-5

M3 - Article

SN - 0925-2312

VL - 55

SP - 151

EP - 167

JO - Neurocomputing

JF - Neurocomputing

IS - 1-2

ER -

SVM regression through variational methods and its sequential implementation

Abstract

Access to Document

Fingerprint

Cite this