Optimization with missing data
Optimization with missing data
Engineering optimization relies routinely on deterministic computer based design evaluations, typically comprising geometry creation, mesh generation and numerical simulation. Simple optimization routines tend to stall and require user intervention if a failure occurs at any of these stages. This motivated us to develop an optimization strategy based on surrogate modelling, which penalizes the likely failure regions of the design space without prior knowledge of their locations. A Gaussian process based design improvement expectation measure guides the search towards the feasible global optimum.
global optimization, imputation, kriging
935-945
Forrester, Alexander I.J.
176bf191-3fc2-46b4-80e0-9d9a0cd7a572
Sobester, Andras
096857b0-cad6-45ae-9ae6-e66b8cc5d81b
Keane, Andy J.
26d7fa33-5415-4910-89d8-fb3620413def
2006
Forrester, Alexander I.J.
176bf191-3fc2-46b4-80e0-9d9a0cd7a572
Sobester, Andras
096857b0-cad6-45ae-9ae6-e66b8cc5d81b
Keane, Andy J.
26d7fa33-5415-4910-89d8-fb3620413def
Forrester, Alexander I.J., Sobester, Andras and Keane, Andy J.
(2006)
Optimization with missing data.
Proceedings of the Royal Society A, 462 (2067), .
(doi:10.1098/rspa.2005.1608).
Abstract
Engineering optimization relies routinely on deterministic computer based design evaluations, typically comprising geometry creation, mesh generation and numerical simulation. Simple optimization routines tend to stall and require user intervention if a failure occurs at any of these stages. This motivated us to develop an optimization strategy based on surrogate modelling, which penalizes the likely failure regions of the design space without prior knowledge of their locations. A Gaussian process based design improvement expectation measure guides the search towards the feasible global optimum.
Text
RSPA20051608p.pdf
- Other
More information
Published date: 2006
Keywords:
global optimization, imputation, kriging
Identifiers
Local EPrints ID: 23505
URI: http://eprints.soton.ac.uk/id/eprint/23505
ISSN: 1364-5021
PURE UUID: a6a1d160-5888-4d5e-90c2-9c0a3eb8e529
Catalogue record
Date deposited: 13 Mar 2006
Last modified: 16 Mar 2024 03:26
Export record
Altmetrics
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics