Cargando…

Sparsifying priors for Bayesian uncertainty quantification in model discovery

We propose a probabilistic model discovery method for identifying ordinary differential equations governing the dynamics of observed multivariate data. Our method is based on the sparse identification of nonlinear dynamics (SINDy) framework, where models are expressed as sparse linear combinations o...

Descripción completa

Detalles Bibliográficos
Autores principales: Hirsh, Seth M., Barajas-Solano, David A., Kutz, J. Nathan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Royal Society 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8864363/
https://www.ncbi.nlm.nih.gov/pubmed/35223066
http://dx.doi.org/10.1098/rsos.211823
_version_ 1784655449044287488
author Hirsh, Seth M.
Barajas-Solano, David A.
Kutz, J. Nathan
author_facet Hirsh, Seth M.
Barajas-Solano, David A.
Kutz, J. Nathan
author_sort Hirsh, Seth M.
collection PubMed
description We propose a probabilistic model discovery method for identifying ordinary differential equations governing the dynamics of observed multivariate data. Our method is based on the sparse identification of nonlinear dynamics (SINDy) framework, where models are expressed as sparse linear combinations of pre-specified candidate functions. Promoting parsimony through sparsity leads to interpretable models that generalize to unknown data. Instead of targeting point estimates of the SINDy coefficients, we estimate these coefficients via sparse Bayesian inference. The resulting method, uncertainty quantification SINDy (UQ-SINDy), quantifies not only the uncertainty in the values of the SINDy coefficients due to observation errors and limited data, but also the probability of inclusion of each candidate function in the linear combination. UQ-SINDy promotes robustness against observation noise and limited data, interpretability (in terms of model selection and inclusion probabilities) and generalization capacity for out-of-sample forecast. Sparse inference for UQ-SINDy employs Markov chain Monte Carlo, and we explore two sparsifying priors: the spike and slab prior, and the regularized horseshoe prior. UQ-SINDy is shown to discover accurate models in the presence of noise and with orders-of-magnitude less data than current model discovery methods, thus providing a transformative method for real-world applications which have limited data.
format Online
Article
Text
id pubmed-8864363
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher The Royal Society
record_format MEDLINE/PubMed
spelling pubmed-88643632022-02-24 Sparsifying priors for Bayesian uncertainty quantification in model discovery Hirsh, Seth M. Barajas-Solano, David A. Kutz, J. Nathan R Soc Open Sci Mathematics We propose a probabilistic model discovery method for identifying ordinary differential equations governing the dynamics of observed multivariate data. Our method is based on the sparse identification of nonlinear dynamics (SINDy) framework, where models are expressed as sparse linear combinations of pre-specified candidate functions. Promoting parsimony through sparsity leads to interpretable models that generalize to unknown data. Instead of targeting point estimates of the SINDy coefficients, we estimate these coefficients via sparse Bayesian inference. The resulting method, uncertainty quantification SINDy (UQ-SINDy), quantifies not only the uncertainty in the values of the SINDy coefficients due to observation errors and limited data, but also the probability of inclusion of each candidate function in the linear combination. UQ-SINDy promotes robustness against observation noise and limited data, interpretability (in terms of model selection and inclusion probabilities) and generalization capacity for out-of-sample forecast. Sparse inference for UQ-SINDy employs Markov chain Monte Carlo, and we explore two sparsifying priors: the spike and slab prior, and the regularized horseshoe prior. UQ-SINDy is shown to discover accurate models in the presence of noise and with orders-of-magnitude less data than current model discovery methods, thus providing a transformative method for real-world applications which have limited data. The Royal Society 2022-02-23 /pmc/articles/PMC8864363/ /pubmed/35223066 http://dx.doi.org/10.1098/rsos.211823 Text en © 2022 The Authors. https://creativecommons.org/licenses/by/4.0/Published by the Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/ (https://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, provided the original author and source are credited.
spellingShingle Mathematics
Hirsh, Seth M.
Barajas-Solano, David A.
Kutz, J. Nathan
Sparsifying priors for Bayesian uncertainty quantification in model discovery
title Sparsifying priors for Bayesian uncertainty quantification in model discovery
title_full Sparsifying priors for Bayesian uncertainty quantification in model discovery
title_fullStr Sparsifying priors for Bayesian uncertainty quantification in model discovery
title_full_unstemmed Sparsifying priors for Bayesian uncertainty quantification in model discovery
title_short Sparsifying priors for Bayesian uncertainty quantification in model discovery
title_sort sparsifying priors for bayesian uncertainty quantification in model discovery
topic Mathematics
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8864363/
https://www.ncbi.nlm.nih.gov/pubmed/35223066
http://dx.doi.org/10.1098/rsos.211823
work_keys_str_mv AT hirshsethm sparsifyingpriorsforbayesianuncertaintyquantificationinmodeldiscovery
AT barajassolanodavida sparsifyingpriorsforbayesianuncertaintyquantificationinmodeldiscovery
AT kutzjnathan sparsifyingpriorsforbayesianuncertaintyquantificationinmodeldiscovery