Bayesian uncertainty quantification for data-driven equation learning.

Autor:	Martina-Perez S; Mathematical Institute, University of Oxford, Oxford, UK., Simpson MJ; School of Mathematical Sciences, Queensland University of Technology, Brisbane, Australia., Baker RE; Mathematical Institute, University of Oxford, Oxford, UK.
Jazyk:	angličtina
Zdroj:	Proceedings. Mathematical, physical, and engineering sciences [Proc Math Phys Eng Sci] 2021 Oct; Vol. 477 (2254), pp. 20210426. Date of Electronic Publication: 2021 Oct 27.
DOI:	10.1098/rspa.2021.0426
Abstrakt:	Equation learning aims to infer differential equation models from data. While a number of studies have shown that differential equation models can be successfully identified when the data are sufficiently detailed and corrupted with relatively small amounts of noise, the relationship between observation noise and uncertainty in the learned differential equation models remains unexplored. We demonstrate that for noisy datasets there exists great variation in both the structure of the learned differential equation models and their parameter values. We explore how to exploit multiple datasets to quantify uncertainty in the learned models, and at the same time draw mechanistic conclusions about the target differential equations. We showcase our results using simulation data from a relatively straightforward agent-based model (ABM) which has a well-characterized partial differential equation description that provides highly accurate predictions of averaged ABM behaviours in relevant regions of parameter space. Our approach combines equation learning methods with Bayesian inference approaches so that a quantification of uncertainty can be given by the posterior parameter distribution of the learned model. (© 2021 The Authors.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu