A hierarchical multilevel markov chain monte carlo algorithm with applications to uncertainty quantification in subsurface flow
- Submitting institution
-
University of Exeter
- Unit of assessment
- 12 - Engineering
- Output identifier
- 1927
- Type
- D - Journal article
- DOI
-
10.1137/130915005
- Title of journal
- SIAM-ASA Journal on Uncertainty Quantification
- Article number
- -
- First page
- 1075
- Volume
- 3
- Issue
- 1
- ISSN
- 2166-2525
- Open access status
- Out of scope for open access requirements
- Month of publication
- November
- Year of publication
- 2015
- 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
-
3
- Research group(s)
-
G - Materials and Manufacturing
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Paper develops the first Multilevel Markov Chain Monte Carlo Method which reduces the cost of Bayesian calibration of expensive models by two orders of magnitude over standard Markov Chain Monte Carlo Methods. This paper was awarded the SIAM SIGEST award, for the best paper in SIAM UQ for four years. It has now been used in a broad range of applications including material science, subsurface flow and image segmentation in medical diagnosis, whilst also being embedded into major open-source codes PyMC3 and MUQ. This foundational research underpinned the award of a UKRI Turing AI Fellowship (£2.6M).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -