GTApprox: surrogate modeling for industrial design
Mathematical Software
2016-09-06 v1 Computational Engineering, Finance, and Science
Machine Learning
Abstract
We describe GTApprox - a new tool for medium-scale surrogate modeling in industrial design. Compared to existing software, GTApprox brings several innovations: a few novel approximation algorithms, several advanced methods of automated model selection, novel options in the form of hints. We demonstrate the efficiency of GTApprox on a large collection of test problems. In addition, we describe several applications of GTApprox to real engineering problems.
Cite
@article{arxiv.1609.01088,
title = {GTApprox: surrogate modeling for industrial design},
author = {Mikhail Belyaev and Evgeny Burnaev and Ermek Kapushev and Maxim Panov and Pavel Prikhodko and Dmitry Vetrov and Dmitry Yarotsky},
journal= {arXiv preprint arXiv:1609.01088},
year = {2016}
}
Comments
31 pages, 11 figures