TY - GEN UR - http://lib.ugent.be/catalog/pug01:4106551 ID - pug01:4106551 LA - eng TI - Dimensionality reduction of optimization problems using variance based sensitivity analysis PY - 2013 PB - 2013 AU - Ito, Keiichi 000111024277 802001511620 972465056754 0000-0003-1364-9477 AU - Dhaene, Tom AB - We propose a new interaction index derived from the computation of Sobol indices. In optimization, interaction index can be used to detect lack of interaction among input parameters. First order interaction indices if they return zero, means that those parameters can be optimized independently holding other parameters constant. Likewise, second order interaction indices can tell if a combination of two parameter can be optimized independently of other parameters. In this way, the original optimization problem may be decomposed into a set of lower dimensional problems which may then be solved independently and in parallel. The interaction indices can potentially be useful in robust optimization as well, since it provides importance measure in minimizing output variances. ER -Download RIS file
00000nam^a2200301^i^4500 | |||
001 | 4106551 | ||
005 | 20161219154803.0 | ||
008 | 130725s2013------------------------eng-- | ||
024 | a 1854/LU-4106551 2 handle | ||
040 | a UGent | ||
245 | a Dimensionality reduction of optimization problems using variance based sensitivity analysis | ||
260 | c 2013 | ||
520 | a We propose a new interaction index derived from the computation of Sobol indices. In optimization, interaction index can be used to detect lack of interaction among input parameters. First order interaction indices if they return zero, means that those parameters can be optimized independently holding other parameters constant. Likewise, second order interaction indices can tell if a combination of two parameter can be optimized independently of other parameters. In this way, the original optimization problem may be decomposed into a set of lower dimensional problems which may then be solved independently and in parallel. The interaction indices can potentially be useful in robust optimization as well, since it provides importance measure in minimizing output variances. | ||
598 | a V | ||
100 | a Ito, Keiichi 0 000111024277 0 802001511620 0 972465056754 0 0000-0003-1364-9477 | ||
700 | a Dhaene, Tom u TW05 0 801000729981 | ||
650 | a Technology and Engineering | ||
653 | a Dimensionality Reduction | ||
653 | a Independence | ||
653 | a Sensitivity Analysis | ||
653 | a Sobol Indices | ||
653 | a Optimization | ||
653 | a Interaction | ||
856 | 3 Full Text u https://biblio.ugent.be/publication/4106551/file/4110020 z [open] y hdmr_opt_sddom.pdf | ||
920 | a miscellanea | ||
852 | x EA b TW05 | ||
922 | a UGENT-EA |
All data below are available with an Open Data Commons Open Database License. You are free to copy, distribute and use the database; to produce works from the database; to modify, transform and build upon the database. As long as you attribute the data sets to the source, publish your adapted database with ODbL license, and keep the dataset open (don't use technical measures such as DRM to restrict access to the database).
The datasets are also available as weekly exports.