Multiphase MCMC sampling for parameter inference in nonlinear ordinary differential equations

Lazarus, A., Husmeier, D. and Papamarkou, T. (2018) Multiphase MCMC sampling for parameter inference in nonlinear ordinary differential equations. Proceedings of Machine Learning Research, 84, pp. 1252-1260.

[img]
Preview
Text
157675.pdf - Published Version
Available under License Creative Commons Attribution.

3MB

Publisher's URL: http://proceedings.mlr.press/v84/lazarus18a.html

Abstract

Traditionally, ODE parameter inference relies on solving the system of ODEs and assessing fit of the estimated signal with the observations. However, nonlinear ODEs often do not permit closed form solutions. Using numerical methods to solve the equations results in prohibitive computational costs, particularly when one adopts a Bayesian approach in sampling parameters from a posterior distribution. With the introduction of gradient matching, we can abandon the need to numerically solve the system of equations. Inherent in these efficient procedures is an introduction of bias to the learning problem as we no longer sample based on the exact likelihood function. This paper presents a multiphase MCMC approach that attempts to close the gap between efficiency and accuracy. By sampling using a surrogate likelihood, we accelerate convergence to the stationary distribution before sampling using the exact likelihood. We demonstrate that this method combines the efficiency of gradient matching and the accuracy of the exact likelihood scheme.

Item Type:Articles
Additional Information:Conference paper presented at International Conference on Artificial Intelligence and Statistics, 9-11 April 2018, Playa Blanca, Lanzarote, Canary Islands.
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Lazarus, Alan and Husmeier, Professor Dirk and Papamarkou, Dr Theodore
Authors: Lazarus, A., Husmeier, D., and Papamarkou, T.
College/School:College of Science and Engineering > School of Mathematics and Statistics > Statistics
Journal Name:Proceedings of Machine Learning Research
Publisher:PMLR
ISSN:1938-7228
Copyright Holders:Copyright © 2018 The Authors
First Published:First published in Proceedings of Machine Learning Research 84: 1252-1260
Publisher Policy:Reproduced under a Creative Commons License
Related URLs:

University Staff: Request a correction | Enlighten Editors: Update this record

Project CodeAward NoProject NamePrincipal InvestigatorFunder's NameFunder RefLead Dept
633291Computational inference in systems biologyDirk HusmeierEngineering and Physical Sciences Research Council (EPSRC)EP/L020319/1M&S - STATISTICS
694461EPSRC Centre for Multiscale soft tissue mechanics with application to heart & cancerRaymond OgdenEngineering and Physical Sciences Research Council (EPSRC)EP/N014642/1M&S - MATHEMATICS