Main Page

This website is under ongoing construction. Since it is a Wiki, you can help by contributing

About

The Surrogate Modeling Toolbox (SUMO-Toolbox) is a Matlab toolbox that builds accurate, surrogate models (also known as metamodels or response surface models) of a given data source (data set, executable, ...). In doing so the toolbox minimizes the number of simulator evaluations since they are usually expensive. The toolbox tries to be as adaptive and flexible as possible, requiring no user input besides some initial configuration.

The toolbox has been designed to be as pluggable and extensible as possible. This means that, allthough the out-of-the-box tools and algorithms are tailored to black box, behavioral modeling (ie. treat the system as a data generating black box that maps inputs onto outputs) there is nothing that prevents a user from plugging in grey/white box model types/modeling algorithms that DO take into account knowledge of the underlying system. So really the toolbox provides a shared platform where different modeling/sampling techniques can be implemented and benchmarked on a wide variety of problems. You can make it as problem specific or problem generic as you want it to be.

The About page contains more information on the history, features and design goals of the toolbox. On the same page you will also find a number of screenshots and movies.

Main Sections

Please use the links below to navigate the site

• General information
• Downloading the toolbox
• System requirements
• Installation Instructions
• Running the toolbox
• Configuration
  • Toolbox configuration
    • General guidelines
  • Simulator configuration
• Extending the toolbox
• Contribute
• Feedback
• Reporting problems
• Useful Links
• Miscellaneous
• Citing the toolbox

License

The toolbox is freely available for personal, academic, non-profit use only. For all other uses explicit written prior permission must be obtained from Prof. Tom Dhaene. For more information see the License terms.

Open Positions

We are constantly looking for highly motivated PhD candidates or post-docs, that have a strong interest in one of our research topics. Our research area is highly multidisciplinary, and requires strong mathematical, physical and computer science knowledge.

Please contact tom.dhaene@ieee.org

See also the FAQ, Known bugs, Terminology and Tips pages for answers to common questions.