Cargando…
Neural parameter calibration for large-scale multiagent models
Computational models have become a powerful tool in the quantitative sciences to understand the behavior of complex systems that evolve in time. However, they often contain a potentially large number of free parameters whose values cannot be obtained from theory but need to be inferred from data. Th...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
National Academy of Sciences
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9963791/ https://www.ncbi.nlm.nih.gov/pubmed/36763529 http://dx.doi.org/10.1073/pnas.2216415120 |
_version_ | 1784896341970780160 |
---|---|
author | Gaskin, Thomas Pavliotis, Grigorios A. Girolami, Mark |
author_facet | Gaskin, Thomas Pavliotis, Grigorios A. Girolami, Mark |
author_sort | Gaskin, Thomas |
collection | PubMed |
description | Computational models have become a powerful tool in the quantitative sciences to understand the behavior of complex systems that evolve in time. However, they often contain a potentially large number of free parameters whose values cannot be obtained from theory but need to be inferred from data. This is especially the case for models in the social sciences, economics, or computational epidemiology. Yet, many current parameter estimation methods are mathematically involved and computationally slow to run. In this paper, we present a computationally simple and fast method to retrieve accurate probability densities for model parameters using neural differential equations. We present a pipeline comprising multiagent models acting as forward solvers for systems of ordinary or stochastic differential equations and a neural network to then extract parameters from the data generated by the model. The two combined create a powerful tool that can quickly estimate densities on model parameters, even for very large systems. We demonstrate the method on synthetic time series data of the SIR model of the spread of infection and perform an in-depth analysis of the Harris–Wilson model of economic activity on a network, representing a nonconvex problem. For the latter, we apply our method both to synthetic data and to data of economic activity across Greater London. We find that our method calibrates the model orders of magnitude more accurately than a previous study of the same dataset using classical techniques, while running between 195 and 390 times faster. |
format | Online Article Text |
id | pubmed-9963791 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | National Academy of Sciences |
record_format | MEDLINE/PubMed |
spelling | pubmed-99637912023-02-26 Neural parameter calibration for large-scale multiagent models Gaskin, Thomas Pavliotis, Grigorios A. Girolami, Mark Proc Natl Acad Sci U S A Physical Sciences Computational models have become a powerful tool in the quantitative sciences to understand the behavior of complex systems that evolve in time. However, they often contain a potentially large number of free parameters whose values cannot be obtained from theory but need to be inferred from data. This is especially the case for models in the social sciences, economics, or computational epidemiology. Yet, many current parameter estimation methods are mathematically involved and computationally slow to run. In this paper, we present a computationally simple and fast method to retrieve accurate probability densities for model parameters using neural differential equations. We present a pipeline comprising multiagent models acting as forward solvers for systems of ordinary or stochastic differential equations and a neural network to then extract parameters from the data generated by the model. The two combined create a powerful tool that can quickly estimate densities on model parameters, even for very large systems. We demonstrate the method on synthetic time series data of the SIR model of the spread of infection and perform an in-depth analysis of the Harris–Wilson model of economic activity on a network, representing a nonconvex problem. For the latter, we apply our method both to synthetic data and to data of economic activity across Greater London. We find that our method calibrates the model orders of magnitude more accurately than a previous study of the same dataset using classical techniques, while running between 195 and 390 times faster. National Academy of Sciences 2023-02-10 2023-02-14 /pmc/articles/PMC9963791/ /pubmed/36763529 http://dx.doi.org/10.1073/pnas.2216415120 Text en Copyright © 2023 the Author(s). Published by PNAS. https://creativecommons.org/licenses/by/4.0/This open access article is distributed under Creative Commons Attribution License 4.0 (CC BY) (https://creativecommons.org/licenses/by/4.0/) . |
spellingShingle | Physical Sciences Gaskin, Thomas Pavliotis, Grigorios A. Girolami, Mark Neural parameter calibration for large-scale multiagent models |
title | Neural parameter calibration for large-scale multiagent models |
title_full | Neural parameter calibration for large-scale multiagent models |
title_fullStr | Neural parameter calibration for large-scale multiagent models |
title_full_unstemmed | Neural parameter calibration for large-scale multiagent models |
title_short | Neural parameter calibration for large-scale multiagent models |
title_sort | neural parameter calibration for large-scale multiagent models |
topic | Physical Sciences |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9963791/ https://www.ncbi.nlm.nih.gov/pubmed/36763529 http://dx.doi.org/10.1073/pnas.2216415120 |
work_keys_str_mv | AT gaskinthomas neuralparametercalibrationforlargescalemultiagentmodels AT pavliotisgrigoriosa neuralparametercalibrationforlargescalemultiagentmodels AT girolamimark neuralparametercalibrationforlargescalemultiagentmodels |