Model guided sampling optimization with gaussian processes for expensive black-box optimization

Autoři: Lukáš Bajer
prof. RNDr. Ing. Martin Holeňa, CSc.
Rok vydání: 2013
Typ publikace: Konferenční příspěvek (mezinárodní konf.)
Periodikum: GECCO Companion '13. Proceeding of the fifteenth annual conference companion on Genetic and evolutionary computation conference companion
Projekty: Učení funkcionálních vztahů z vysoce dimenzionálních dat
Konstrukce pokročilých srozumitelných klasifikátorů
Bibliografická citace: BAJER, Lukáš, CHARYPAR, V., HOLEŇA, Martin. Model guided sampling optimization with gaussian processes for expensive black-box optimization. In: BLUM, C., ed. GECCO Companion '13. Proceeding of the fifteenth annual conference companion on Genetic and evolutionary computation conference companion. New York: ACM, 2013, S. 1715-1716. ISBN 978-1-4503-1964-5
Dostupné z: doi:10.1145/2464576.2480794
DOI: 10.1145/2464576.2480794
ID: 394260

Abstrakt

Model Guided Sampling Optimization (MGSO) is a novel expensive black-box optimization method based on a combination of ideas from Estimation of Distribution Algorithms and global optimization methods using Gaussian Processes. The algorithm is described and its implementation tested on three benchmark functions as a proof of concept.