Using machine learning as a surrogate model for agent-based simulations.
In this proof-of-concept work, we evaluate the performance of multiple machine-learning methods as surrogate models for use in the analysis of agent-based models (ABMs). Analysing agent-based modelling outputs can be challenging, as the relationships between input parameters can be non-linear or eve...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2022-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0263150&type=printable |
| _version_ | 1826804380618194944 |
|---|---|
| author | Claudio Angione Eric Silverman Elisabeth Yaneske |
| author_facet | Claudio Angione Eric Silverman Elisabeth Yaneske |
| author_sort | Claudio Angione |
| collection | DOAJ |
| description | In this proof-of-concept work, we evaluate the performance of multiple machine-learning methods as surrogate models for use in the analysis of agent-based models (ABMs). Analysing agent-based modelling outputs can be challenging, as the relationships between input parameters can be non-linear or even chaotic even in relatively simple models, and each model run can require significant CPU time. Surrogate modelling, in which a statistical model of the ABM is constructed to facilitate detailed model analyses, has been proposed as an alternative to computationally costly Monte Carlo methods. Here we compare multiple machine-learning methods for ABM surrogate modelling in order to determine the approaches best suited as a surrogate for modelling the complex behaviour of ABMs. Our results suggest that, in most scenarios, artificial neural networks (ANNs) and gradient-boosted trees outperform Gaussian process surrogates, currently the most commonly used method for the surrogate modelling of complex computational models. ANNs produced the most accurate model replications in scenarios with high numbers of model runs, although training times were longer than the other methods. We propose that agent-based modelling would benefit from using machine-learning methods for surrogate modelling, as this can facilitate more robust sensitivity analyses for the models while also reducing CPU time consumption when calibrating and analysing the simulation. |
| first_indexed | 2024-04-11T14:19:53Z |
| format | Article |
| id | doaj.art-4bb0ddfb7e3041f2b61a468a8c294de8 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2025-03-17T01:53:38Z |
| publishDate | 2022-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-4bb0ddfb7e3041f2b61a468a8c294de82025-02-15T05:30:48ZengPublic Library of Science (PLoS)PLoS ONE1932-62032022-01-01172e026315010.1371/journal.pone.0263150Using machine learning as a surrogate model for agent-based simulations.Claudio AngioneEric SilvermanElisabeth YaneskeIn this proof-of-concept work, we evaluate the performance of multiple machine-learning methods as surrogate models for use in the analysis of agent-based models (ABMs). Analysing agent-based modelling outputs can be challenging, as the relationships between input parameters can be non-linear or even chaotic even in relatively simple models, and each model run can require significant CPU time. Surrogate modelling, in which a statistical model of the ABM is constructed to facilitate detailed model analyses, has been proposed as an alternative to computationally costly Monte Carlo methods. Here we compare multiple machine-learning methods for ABM surrogate modelling in order to determine the approaches best suited as a surrogate for modelling the complex behaviour of ABMs. Our results suggest that, in most scenarios, artificial neural networks (ANNs) and gradient-boosted trees outperform Gaussian process surrogates, currently the most commonly used method for the surrogate modelling of complex computational models. ANNs produced the most accurate model replications in scenarios with high numbers of model runs, although training times were longer than the other methods. We propose that agent-based modelling would benefit from using machine-learning methods for surrogate modelling, as this can facilitate more robust sensitivity analyses for the models while also reducing CPU time consumption when calibrating and analysing the simulation.https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0263150&type=printable |
| spellingShingle | Claudio Angione Eric Silverman Elisabeth Yaneske Using machine learning as a surrogate model for agent-based simulations. PLoS ONE |
| title | Using machine learning as a surrogate model for agent-based simulations. |
| title_full | Using machine learning as a surrogate model for agent-based simulations. |
| title_fullStr | Using machine learning as a surrogate model for agent-based simulations. |
| title_full_unstemmed | Using machine learning as a surrogate model for agent-based simulations. |
| title_short | Using machine learning as a surrogate model for agent-based simulations. |
| title_sort | using machine learning as a surrogate model for agent based simulations |
| url | https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0263150&type=printable |
| work_keys_str_mv | AT claudioangione usingmachinelearningasasurrogatemodelforagentbasedsimulations AT ericsilverman usingmachinelearningasasurrogatemodelforagentbasedsimulations AT elisabethyaneske usingmachinelearningasasurrogatemodelforagentbasedsimulations |