000012468 001__ 12468 000012468 005__ 20160816141447.0 000012468 04107 $$aeng 000012468 046__ $$k2016-05-09 000012468 100__ $$aMyšáková, E. 000012468 24500 $$aPRESS WEIGHTED AVERAGE SURROGATE: TRIAL TESTS IN 2D 000012468 24630 $$n22.$$pEngineering Mechanics 2016 000012468 260__ $$bInstitute of Thermomechanics AS CR, v.v.i., Praha 000012468 506__ $$arestricted 000012468 520__ $$2eng$$aSurrogate modeling (Meta-modeling) is an often used tool for analysis of behavior of complex systems which are usually described by computationally demanding models. Surrogate models provide an approximation of the original model's response in a fraction of time and therefore are suitable when multiple evaluations are needed. Many types of meta-models exist and each suits another type of problem. On the other hand it is not always possible to select the right meta-model in advance. Therefore parallel construction of several meta-models and their subsequent comparison and combining can be utilized with advantage. A typical method called PRESS weighted average surrogate which uses the prediction sum of squares obtained by cross-validation for computation of the weights for linear combination of individual surrogates is discussed in this contribution and illustrated on several 2-dimensional benchmark examples using a group of different meta-models. 000012468 540__ $$aText je chráněný podle autorského zákona č. 121/2000 Sb. 000012468 653__ $$aMeta-modeling, Ensemble of Surrogates, Cross-validation, Root Mean Square Error, Prediction Sum of Squares. 000012468 7112_ $$aEngineering Mechanics 2016$$cSvratka, CZ$$d2016-05-09 / 2016-05-12$$gEM2016 000012468 720__ $$aMyšáková, E.$$iLepš, M. 000012468 8560_ $$ffischerc@itam.cas.cz 000012468 8564_ $$s206194$$uhttps://invenio.itam.cas.cz/record/12468/files/098bo_o_sol.pdf$$yOriginal version of the author's contribution as presented on CD, id 098, section SOL. 000012468 962__ $$r12369 000012468 980__ $$aPAPER