Bayesian Optimization for Likelihood-Free Inference of Simulator-Based Statistical Models
- Submitting institution
-
University of Edinburgh
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 58833454
- Type
- D - Journal article
- DOI
-
-
- Title of journal
- Journal of Machine Learning Research
- Article number
- -
- First page
- 1
- Volume
- 17
- Issue
- 125
- ISSN
- 1532-4435
- Open access status
- Out of scope for open access requirements
- Month of publication
- August
- Year of publication
- 2016
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
1
- Research group(s)
-
B - Data Science and Artificial Intelligence
- Citation count
- 59
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- First paper to combine probabilistic modelling and decision making under uncertainty to accelerate Bayesian inference of intractable simulator-based models (approximate Bayesian computation / likelihood-free inference). The method reduces the computation cost by factors of 1000 or more compared to state-of-the art methods. Enabled inference for models that were previously out of reach (e.g. in the health sciences, Corander et al, 2017, Nature Ecology & Evolution), was applied in a wide range of disciplines (e.g. material science: npj Comp Mat 5(35), 2019, or astrophysics: Phys. Rev. D 98, 063511) and underpinned the development of new statistical inference software (ELFI, JMLR, 2017).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -