Surrogate models are a common tool to perform expensive optimization or uncertainty analysis in cost-effective way. The structure and the application of surrogate is well known, but there still potential for improvement. model can be divided into three parts, which considered separately. sampling plan, black box selection itself. plan presented on basis two case studies. In plans, example FOSS ...

Abstract We present an introduction to some of the state art in reduced order and surrogate modeling gravitational-wave (GW) science. Approaches that we cover include principal component analysis, proper orthogonal (singular value) decompositions, basis approach, empirical interpolation method, quadratures, compressed likelihood evaluations. divide review into three parts: representation/compre...

Neural network surrogate models are often used to replace complex mathematical in black-box and grey-box optimization. This strategy essentially uses samples generated from a model fit data-driven, reduced-order more amenable for can be trained Sobolev spaces, i.e., match the function not only terms of output values, but also values their derivatives arbitrary degree. paper examines direct impa...