GTApprox: surrogate modeling for industrial design

Belyaev, Mikhail, Burnaev, Evgeny, Kapushev, Ermek, Panov, Maxim, Prikhodko, Pavel, Vetrov, Dmitry, Yarotsky, Dmitry

Sep-5-2016–arXiv.org Machine Learning

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.

neural network, optimization problem, surrogate model, (18 more...)

arXiv.org Machine Learning

Sep-5-2016

arXiv.org PDF

Add feedback

Country:
- Europe
  - Germany (0.14)
  - Russia (0.14)
  - Spain (0.14)
- North America > United States (0.14)

Genre:
- Instructional Material > Course Syllabus & Notes (0.46)
- Research Report (1.00)

Industry:
- Aerospace & Defense (0.46)

Technology:
- Information Technology
  - Artificial Intelligence
    - Machine Learning
      - Neural Networks (0.93)
      - Statistical Learning > Regression (1.00)
    - Representation & Reasoning > Optimization (1.00)
  - Data Science (1.00)

Duplicate Docs Excel Report

Title
None found

Similar Docs Excel Report more

Title	Similarity	Source
None found