A new variational radial basis function approximation for inference in multivariate diffusions

Michail D. Vrettas, Dan Cornford, Manfred Opper, Yuan Shen

Research output: Contribution to journalArticle

Abstract

In this paper we present a radial basis function based extension to a recently proposed variational algorithm for approximate inference for diffusion processes. Inference, for state and in particular (hyper-) parameters, in diffusion processes is a challenging and crucial task. We show that the new radial basis function approximation based algorithm converges to the original algorithm and has beneficial characteristics when estimating (hyper-)parameters. We validate our new approach on a nonlinear double well potential dynamical system.
Original languageEnglish
Pages (from-to)1186-1198
Number of pages13
JournalNeurocomputing
Volume73
Issue number7-9
DOIs
Publication statusPublished - Mar 2010
Event17th European Symposium on Artificial Neural Networks: Advances in Computational Intelligence and Learning - Bruges, Belgium
Duration: 22 Apr 200924 Apr 2009

Fingerprint

Dynamical systems

Bibliographical note

NOTICE: this is the author’s version of a work that was accepted for publication in Neurocomputing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Vrettas, Michail D.; Cornford, Dan; Opper, Manfred and Shen, Yuan (2010). A variational basis function approximation for diffusion processes. Neurocomputing, 73 (7-9), pp. 1186-1198. DOI 10.1016/j.neucom.2009.11.026

Keywords

  • radial basis functions
  • dynamical systems
  • stochastic differential equations
  • parameter estimation
  • Bayesian inference

Cite this

Vrettas, Michail D. ; Cornford, Dan ; Opper, Manfred ; Shen, Yuan. / A new variational radial basis function approximation for inference in multivariate diffusions. In: Neurocomputing. 2010 ; Vol. 73, No. 7-9. pp. 1186-1198.
@article{bc40141be60e43b2a0b13527f542a1a5,
title = "A new variational radial basis function approximation for inference in multivariate diffusions",
abstract = "In this paper we present a radial basis function based extension to a recently proposed variational algorithm for approximate inference for diffusion processes. Inference, for state and in particular (hyper-) parameters, in diffusion processes is a challenging and crucial task. We show that the new radial basis function approximation based algorithm converges to the original algorithm and has beneficial characteristics when estimating (hyper-)parameters. We validate our new approach on a nonlinear double well potential dynamical system.",
keywords = "radial basis functions, dynamical systems, stochastic differential equations, parameter estimation, Bayesian inference",
author = "Vrettas, {Michail D.} and Dan Cornford and Manfred Opper and Yuan Shen",
note = "NOTICE: this is the author’s version of a work that was accepted for publication in Neurocomputing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Vrettas, Michail D.; Cornford, Dan; Opper, Manfred and Shen, Yuan (2010). A variational basis function approximation for diffusion processes. Neurocomputing, 73 (7-9), pp. 1186-1198. DOI 10.1016/j.neucom.2009.11.026",
year = "2010",
month = "3",
doi = "10.1016/j.neucom.2009.11.026",
language = "English",
volume = "73",
pages = "1186--1198",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",
number = "7-9",

}

A new variational radial basis function approximation for inference in multivariate diffusions. / Vrettas, Michail D.; Cornford, Dan; Opper, Manfred; Shen, Yuan.

In: Neurocomputing, Vol. 73, No. 7-9, 03.2010, p. 1186-1198.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A new variational radial basis function approximation for inference in multivariate diffusions

AU - Vrettas, Michail D.

AU - Cornford, Dan

AU - Opper, Manfred

AU - Shen, Yuan

N1 - NOTICE: this is the author’s version of a work that was accepted for publication in Neurocomputing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Vrettas, Michail D.; Cornford, Dan; Opper, Manfred and Shen, Yuan (2010). A variational basis function approximation for diffusion processes. Neurocomputing, 73 (7-9), pp. 1186-1198. DOI 10.1016/j.neucom.2009.11.026

PY - 2010/3

Y1 - 2010/3

N2 - In this paper we present a radial basis function based extension to a recently proposed variational algorithm for approximate inference for diffusion processes. Inference, for state and in particular (hyper-) parameters, in diffusion processes is a challenging and crucial task. We show that the new radial basis function approximation based algorithm converges to the original algorithm and has beneficial characteristics when estimating (hyper-)parameters. We validate our new approach on a nonlinear double well potential dynamical system.

AB - In this paper we present a radial basis function based extension to a recently proposed variational algorithm for approximate inference for diffusion processes. Inference, for state and in particular (hyper-) parameters, in diffusion processes is a challenging and crucial task. We show that the new radial basis function approximation based algorithm converges to the original algorithm and has beneficial characteristics when estimating (hyper-)parameters. We validate our new approach on a nonlinear double well potential dynamical system.

KW - radial basis functions

KW - dynamical systems

KW - stochastic differential equations

KW - parameter estimation

KW - Bayesian inference

UR - http://www.scopus.com/inward/record.url?scp=77649234513&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2009.11.026

DO - 10.1016/j.neucom.2009.11.026

M3 - Article

VL - 73

SP - 1186

EP - 1198

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

IS - 7-9

ER -