An adjoint for likelihood maximization
An adjoint for likelihood maximization
The process of likelihood maximization can be found in many different areas of computational modelling. However, the construction of such models via likelihood maximization requires the solution of a difficult multi-modal optimization problem involving an expensive O(n3) factorization. The optimization techniques used to solve this problem may require many such factorizations and can result in a significant bottle-neck. This article derives an adjoint formulation of the likelihood employed in the construction of a kriging model via reverse algorithmic differentiation. This adjoint is found to calculate the likelihood and all of its derivatives more efficiently than the standard analytical method and can therefore be utilised within a simple local search or within a hybrid global optimization to accelerate convergence and therefore reduce the cost of the likelihood optimization.
kriging, agorithmic differentiation, likelihood maximization
3267-3287
Toal, David J.J.
dc67543d-69d2-4f27-a469-42195fa31a68
Forrester, Alexander I.J.
176bf191-3fc2-46b4-80e0-9d9a0cd7a572
Bressloff, Neil W.
4f531e64-dbb3-41e3-a5d3-e6a5a7a77c92
Keane, Andy J.
26d7fa33-5415-4910-89d8-fb3620413def
Holden, Carren
c4135ae9-a4c9-42d9-88db-7b45a8d73a78
November 2009
Toal, David J.J.
dc67543d-69d2-4f27-a469-42195fa31a68
Forrester, Alexander I.J.
176bf191-3fc2-46b4-80e0-9d9a0cd7a572
Bressloff, Neil W.
4f531e64-dbb3-41e3-a5d3-e6a5a7a77c92
Keane, Andy J.
26d7fa33-5415-4910-89d8-fb3620413def
Holden, Carren
c4135ae9-a4c9-42d9-88db-7b45a8d73a78
Toal, David J.J., Forrester, Alexander I.J., Bressloff, Neil W., Keane, Andy J. and Holden, Carren
(2009)
An adjoint for likelihood maximization.
Proceedings of the Royal Society A, 465 (2111), .
(doi:10.1098/rspa.2009.0096).
Abstract
The process of likelihood maximization can be found in many different areas of computational modelling. However, the construction of such models via likelihood maximization requires the solution of a difficult multi-modal optimization problem involving an expensive O(n3) factorization. The optimization techniques used to solve this problem may require many such factorizations and can result in a significant bottle-neck. This article derives an adjoint formulation of the likelihood employed in the construction of a kriging model via reverse algorithmic differentiation. This adjoint is found to calculate the likelihood and all of its derivatives more efficiently than the standard analytical method and can therefore be utilised within a simple local search or within a hybrid global optimization to accelerate convergence and therefore reduce the cost of the likelihood optimization.
Text
Toal_09.pdf
- Accepted Manuscript
More information
Published date: November 2009
Keywords:
kriging, agorithmic differentiation, likelihood maximization
Organisations:
Computational Engineering and Design
Identifiers
Local EPrints ID: 71661
URI: http://eprints.soton.ac.uk/id/eprint/71661
ISSN: 1364-5021
PURE UUID: 8bdc113d-384b-4db5-a7a3-a7626bbb3c64
Catalogue record
Date deposited: 17 Dec 2009
Last modified: 14 Mar 2024 02:53
Export record
Altmetrics
Contributors
Author:
Carren Holden
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
