Multifidelity approximate Bayesian computation with sequential Monte Carlo parameter sampling
Multifidelity approximate Bayesian computation (MF-ABC) is a likelihood-free technique for parameter inference that exploits model approximations to significantly increase the speed of ABC algorithms (Prescott and Baker, 2020). Previous work has considered MF-ABC only in the context of rejection sam...
| Main Authors: | , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Society for Industrial and Applied Mathematics
2021
|
| _version_ | 1826295085268992000 |
|---|---|
| author | Prescott, TP Baker, RE |
| author_facet | Prescott, TP Baker, RE |
| author_sort | Prescott, TP |
| collection | OXFORD |
| description | Multifidelity approximate Bayesian computation (MF-ABC) is a likelihood-free
technique for parameter inference that exploits model approximations to
significantly increase the speed of ABC algorithms (Prescott and Baker, 2020).
Previous work has considered MF-ABC only in the context of rejection sampling,
which does not explore parameter space particularly efficiently. In this work,
we integrate the multifidelity approach with the ABC sequential Monte Carlo
(ABC-SMC) algorithm into a new MF-ABC-SMC algorithm. We show that the
improvements generated by each of ABC-SMC and MF-ABC to the efficiency of
generating Monte Carlo samples and estimates from the ABC posterior are
amplified when the two techniques are used together. |
| first_indexed | 2024-03-07T03:55:40Z |
| format | Journal article |
| id | oxford-uuid:c2cb8307-5fbd-4a70-994b-d1e6328f881c |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T03:55:40Z |
| publishDate | 2021 |
| publisher | Society for Industrial and Applied Mathematics |
| record_format | dspace |
| spelling | oxford-uuid:c2cb8307-5fbd-4a70-994b-d1e6328f881c2022-03-27T06:11:31ZMultifidelity approximate Bayesian computation with sequential Monte Carlo parameter samplingJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:c2cb8307-5fbd-4a70-994b-d1e6328f881cEnglishSymplectic ElementsSociety for Industrial and Applied Mathematics2021Prescott, TPBaker, REMultifidelity approximate Bayesian computation (MF-ABC) is a likelihood-free technique for parameter inference that exploits model approximations to significantly increase the speed of ABC algorithms (Prescott and Baker, 2020). Previous work has considered MF-ABC only in the context of rejection sampling, which does not explore parameter space particularly efficiently. In this work, we integrate the multifidelity approach with the ABC sequential Monte Carlo (ABC-SMC) algorithm into a new MF-ABC-SMC algorithm. We show that the improvements generated by each of ABC-SMC and MF-ABC to the efficiency of generating Monte Carlo samples and estimates from the ABC posterior are amplified when the two techniques are used together. |
| spellingShingle | Prescott, TP Baker, RE Multifidelity approximate Bayesian computation with sequential Monte Carlo parameter sampling |
| title | Multifidelity approximate Bayesian computation with sequential Monte Carlo parameter sampling |
| title_full | Multifidelity approximate Bayesian computation with sequential Monte Carlo parameter sampling |
| title_fullStr | Multifidelity approximate Bayesian computation with sequential Monte Carlo parameter sampling |
| title_full_unstemmed | Multifidelity approximate Bayesian computation with sequential Monte Carlo parameter sampling |
| title_short | Multifidelity approximate Bayesian computation with sequential Monte Carlo parameter sampling |
| title_sort | multifidelity approximate bayesian computation with sequential monte carlo parameter sampling |
| work_keys_str_mv | AT prescotttp multifidelityapproximatebayesiancomputationwithsequentialmontecarloparametersampling AT bakerre multifidelityapproximatebayesiancomputationwithsequentialmontecarloparametersampling |