Global optimization based on noisy evaluations: an empirical study of two statistical approaches

Abstract : The optimization of the output of complex computer codes has often to be achieved with a small budget of evaluations. Algorithms dedicated to such problems have been developed and compared, such as the Expected Improvement algorithm (EI) or the Informational Approach to Global Optimization (IAGO). However, the influence of noisy evaluation results on the outcome of these comparisons has often been neglected, despite its frequent appearance in industrial problems. In this paper, empirical convergence rates for EI and IAGO are compared when an additive noise corrupts the result of an evaluation. IAGO appears more efficient than EI and various modifications of EI designed to deal with noisy evaluations.
Document type :
Journal articles
Complete list of metadatas

https://hal-supelec.archives-ouvertes.fr/hal-00354656
Contributor : Karine El Rassi <>
Submitted on : Tuesday, January 20, 2009 - 3:06:43 PM
Last modification on : Thursday, April 5, 2018 - 12:30:04 PM
Long-term archiving on : Tuesday, June 8, 2010 - 5:44:25 PM

File

paper_submitted_v2.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Emmanuel Vazquez, Julien Villemonteix, Maryan Sidorkiewicz, Eric Walter. Global optimization based on noisy evaluations: an empirical study of two statistical approaches. Journal of Global Optimization, Springer Verlag, 2008, Vol. 43 ((2-3)), pp. 373-389. ⟨10.1007/s10898-008-9313-y⟩. ⟨hal-00354656v1⟩

Share

Metrics

Record views

41

Files downloads

28