Laboratory of Structural Methods of Data Analysis in Predictive
Modeling Moscow Institute of Physics and Technology
Approximation of a multidimensional dependency based on a linear expansion in a dictionary of parametric functions
We consider the problem of a multidimensional function approximation using a ?nite set of pairs “point”—“function value at this point”. As a model for the function we use expansion in a dictionary containing non-linear parametric functions. Several sub-problems should be solve when constructing an approximation based on such model: extraction of a validation sample, initialization of parameters of the functions from the dictionary, tuning of these parameters. We propose e?cient methods for solving these sub-problems. E?ciency of the proposed approach is demonstrated on some problems of engineering design.

Авторы: Belyaev Mikhail , Burnaev Evgeny

Дата: 17 ноября 2014

Статус: опубликована

Журнал: Informatics and Applications

Том: 7

Выпуск: 3

Страницы: 114-125

Год: 2013

Google scholar:

Направления исследований

Statistical methods

Дополнительные материалы